Google: Gemma 4 31B

Gemma 4 31B Instruct is Google DeepMind's 30.7B dense multimodal model supporting text and image input with text output. Features a 256K token context window, configurable thinking/reasoning mode, na…

google/gemma-4-31b-it-20260402

Gemma 4 31B Instruct is Google DeepMind's 30.7B dense multimodal model supporting text and image input with text output. Features a 256K token context window, configurable thinking/reasoning mode, native function calling, and multilingual support across 140+ languages. Strong on coding, reasoning, and document understanding tasks. Apache 2.0 license.

Specifications & capabilities

Input
image, text, video
Output
text
Context window · tokens
262144
Published model ID
gemma4@gemma-4-31b-it
Capabilities
multimodal, promptCaching, reasoning, structuredOutputs, toolCalling, visionInput
Description source ↗

Evaluations

21–40 / 44

OpenRouter Benchmarks

71.667%

source id
openRouterBenchmarks
leaderboard name
tau2-bench Verified Airline · Accuracy · DekaLLM
dimension
agentic
metric
tau2-bench Verified Airline · Accuracy
rank
8
sample size
50
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
1
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
tau_bench_verified_airline
name
tau2-bench Verified Airline
record_kind
aggregate
provider_name
DekaLLM
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-31T20:19:18.995Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
tau2-bench Verified Airline · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
50
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/tau2-bench-airline
measurements
accuracy
0.716667
cost_per_task_usd
0.01519146
output_tokens_per_task
11470.58
latency_seconds_per_task
819.8232399999999
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
10
benchmark_evaluation
true
openrouter_permaslug
google/gemma-4-31b-it-20260402
source model id
google/gemma-4-31b-it-20260402
Original result ↗

OpenRouter Benchmarks

84.123%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · PrimeIntellect
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
9
sample size
190
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
1
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
gpqa_diamond
name
GPQA Diamond
record_kind
aggregate
provider_name
PrimeIntellect
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-11-01T18:00:07.350Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
GPQA Diamond · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
190
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.841228
cost_per_task_usd
0.0038195157894736842
output_tokens_per_task
17024.936842105264
latency_seconds_per_task
400.7756
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
22
benchmark_evaluation
true
openrouter_permaslug
google/gemma-4-31b-it-20260402
source model id
google/gemma-4-31b-it-20260402
Original result ↗

OpenRouter Benchmarks

77.094%

source id
openRouterBenchmarks
leaderboard name
tau2-bench Verified Airline · Accuracy · DeepInfra
dimension
agentic
metric
tau2-bench Verified Airline · Accuracy
rank
9
sample size
748
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
15
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
tau_bench_verified_airline
name
tau2-bench Verified Airline
record_kind
aggregate
provider_name
DeepInfra
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-31T19:33:25.394Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
tau2-bench Verified Airline · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
748
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/tau2-bench-airline
measurements
accuracy
0.7709447032085562
accuracy_std_dev
0.032091449469013114
cost_per_task_usd
0.03325723529411764
output_tokens_per_task
12250.7807486631
latency_seconds_per_task
455.90675534759356
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
63
benchmark_evaluation
true
openrouter_permaslug
google/gemma-4-31b-it-20260402
source model id
google/gemma-4-31b-it-20260402
Original result ↗

OpenRouter Benchmarks

74.571%

source id
openRouterBenchmarks
leaderboard name
tau2-bench Verified Airline · Accuracy · CoreWeave
dimension
agentic
metric
tau2-bench Verified Airline · Accuracy
rank
10
sample size
350
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
7
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
tau_bench_verified_airline
name
tau2-bench Verified Airline
record_kind
aggregate
provider_name
CoreWeave
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-31T19:15:14.763Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
tau2-bench Verified Airline · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
350
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/tau2-bench-airline
measurements
accuracy
0.7457142857142857
accuracy_std_dev
0.02770085585430936
cost_per_task_usd
0.025092839999999998
output_tokens_per_task
10724.948571428571
latency_seconds_per_task
285.7090542857143
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
22
benchmark_evaluation
true
openrouter_permaslug
google/gemma-4-31b-it-20260402
source model id
google/gemma-4-31b-it-20260402
Original result ↗

OpenRouter Benchmarks

65.632%

source id
openRouterBenchmarks
leaderboard name
tau2-bench Verified Airline · Accuracy · Crusoe
dimension
agentic
metric
tau2-bench Verified Airline · Accuracy
rank
10
sample size
145
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
3
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
tau_bench_verified_airline
name
tau2-bench Verified Airline
record_kind
aggregate
provider_name
Crusoe
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-31T19:23:32.973Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
tau2-bench Verified Airline · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
145
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/tau2-bench-airline
measurements
accuracy
0.6563219310344829
accuracy_std_dev
0.17320901072467976
cost_per_task_usd
0.02278490344827586
output_tokens_per_task
8725.041379310345
latency_seconds_per_task
272.4172068965517
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
13
benchmark_evaluation
true
openrouter_permaslug
google/gemma-4-31b-it-20260402
source model id
google/gemma-4-31b-it-20260402
Original result ↗

OpenRouter Benchmarks

73.667%

source id
openRouterBenchmarks
leaderboard name
tau2-bench Verified Airline · Accuracy · Phala
dimension
agentic
metric
tau2-bench Verified Airline · Accuracy
rank
11
sample size
100
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
2
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
tau_bench_verified_airline
name
tau2-bench Verified Airline
record_kind
aggregate
provider_name
Phala
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-08-29T16:43:42.382Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
tau2-bench Verified Airline · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
100
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/tau2-bench-airline
measurements
accuracy
0.7366665
accuracy_std_dev
0.0033334999999999892
cost_per_task_usd
0.020005850000000002
output_tokens_per_task
11689.47
latency_seconds_per_task
329.58574
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
23
benchmark_evaluation
true
openrouter_permaslug
google/gemma-4-31b-it-20260402
source model id
google/gemma-4-31b-it-20260402
Original result ↗

OpenRouter Benchmarks

42.333%

source id
openRouterBenchmarks
leaderboard name
tau2-bench Verified Airline · Accuracy · Modal
dimension
agentic
metric
tau2-bench Verified Airline · Accuracy
rank
11
sample size
50
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
1
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
tau_bench_verified_airline
name
tau2-bench Verified Airline
record_kind
aggregate
provider_name
Modal
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-11-04T09:13:34.415Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
tau2-bench Verified Airline · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
50
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/tau2-bench-airline
measurements
accuracy
0.423333
cost_per_task_usd
0.00707156
output_tokens_per_task
3450.68
latency_seconds_per_task
19.040860000000002
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
11
benchmark_evaluation
true
openrouter_permaslug
google/gemma-4-31b-it-20260402
source model id
google/gemma-4-31b-it-20260402
Original result ↗

OpenRouter Benchmarks

85.494%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · SiliconFlow
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
13
sample size
1188
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
6
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
gpqa_diamond
name
GPQA Diamond
record_kind
aggregate
provider_name
SiliconFlow
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-31T16:05:36.294Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
GPQA Diamond · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
1188
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.8549381666666667
accuracy_std_dev
0.008098201002211657
cost_per_task_usd
0.015701257575757577
output_tokens_per_task
19145.29797979798
latency_seconds_per_task
717.1981986531987
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
34
benchmark_evaluation
true
openrouter_permaslug
google/gemma-4-31b-it-20260402
source model id
google/gemma-4-31b-it-20260402
Original result ↗

OpenRouter Benchmarks

84.43%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · Phala
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
13
sample size
395
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
2
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
gpqa_diamond
name
GPQA Diamond
record_kind
aggregate
provider_name
Phala
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-08-29T19:08:46.030Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
GPQA Diamond · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
395
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.844303929113924
accuracy_std_dev
0.0093105
cost_per_task_usd
0.008944994936708861
output_tokens_per_task
19303.701265822783
latency_seconds_per_task
480.1120911392405
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
24
benchmark_evaluation
true
openrouter_permaslug
google/gemma-4-31b-it-20260402
source model id
google/gemma-4-31b-it-20260402
Original result ↗

OpenRouter Benchmarks

37.872%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · Io Net
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
15
sample size
224
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
2
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
gpqa_diamond
name
GPQA Diamond
record_kind
aggregate
provider_name
Io Net
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-11-11T05:27:07.237Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
GPQA Diamond · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
224
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.3787201875
accuracy_std_dev
0.3723545
cost_per_task_usd
0.0059331517857142855
output_tokens_per_task
5127.165178571428
latency_seconds_per_task
145.88342857142857
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
15
benchmark_evaluation
true
openrouter_permaslug
google/gemma-4-31b-it-20260402
source model id
google/gemma-4-31b-it-20260402
Original result ↗

OpenRouter Benchmarks

85.006%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · Parasail
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
16
sample size
1375
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
7
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
gpqa_diamond
name
GPQA Diamond
record_kind
aggregate
provider_name
Parasail
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-31T16:12:28.235Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
GPQA Diamond · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
1375
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.850060552
accuracy_std_dev
0.01012123337849893
cost_per_task_usd
0.0072516559999999996
output_tokens_per_task
17945.874909090908
latency_seconds_per_task
627.5111978181818
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
36
benchmark_evaluation
true
openrouter_permaslug
google/gemma-4-31b-it-20260402
source model id
google/gemma-4-31b-it-20260402
Original result ↗

OpenRouter Benchmarks

82.203%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · CoreWeave
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
17
sample size
1386
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
7
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
gpqa_diamond
name
GPQA Diamond
record_kind
aggregate
provider_name
CoreWeave
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-31T16:12:28.233Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
GPQA Diamond · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
1386
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.8220297142857144
accuracy_std_dev
0.020075599465977787
cost_per_task_usd
0.006425138528138528
output_tokens_per_task
18776.16810966811
latency_seconds_per_task
406.7264509379509
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
27
benchmark_evaluation
true
openrouter_permaslug
google/gemma-4-31b-it-20260402
source model id
google/gemma-4-31b-it-20260402
Original result ↗

OpenRouter Benchmarks

83.8%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · Together
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
21
sample size
571
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
3
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
gpqa_diamond
name
GPQA Diamond
record_kind
aggregate
provider_name
Together
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-08-29T18:29:51.235Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
GPQA Diamond · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
571
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.8380034220665499
accuracy_std_dev
0.010074705190062243
cost_per_task_usd
0.01617205078809107
output_tokens_per_task
17113.875656742555
latency_seconds_per_task
843.9188598949212
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
25
benchmark_evaluation
true
openrouter_permaslug
google/gemma-4-31b-it-20260402
source model id
google/gemma-4-31b-it-20260402
Original result ↗

OpenRouter Benchmarks

65.733%

source id
openRouterBenchmarks
leaderboard name
tau2-bench Verified Airline · Accuracy · SiliconFlow
dimension
agentic
metric
tau2-bench Verified Airline · Accuracy
rank
21
sample size
250
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
5
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
tau_bench_verified_airline
name
tau2-bench Verified Airline
record_kind
aggregate
provider_name
SiliconFlow
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-31T19:32:40.676Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
tau2-bench Verified Airline · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
250
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/tau2-bench-airline
measurements
accuracy
0.6573334
accuracy_std_dev
0.1373785957776538
cost_per_task_usd
0.06366923599999999
output_tokens_per_task
10164.08
latency_seconds_per_task
573.529892
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
28
benchmark_evaluation
true
openrouter_permaslug
google/gemma-4-31b-it-20260402
source model id
google/gemma-4-31b-it-20260402
Original result ↗

OpenRouter Benchmarks

84.403%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · Novita
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
23
sample size
1293
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
7
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
gpqa_diamond
name
GPQA Diamond
record_kind
aggregate
provider_name
Novita
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-31T16:34:27.227Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
GPQA Diamond · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
1293
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.844032090487239
accuracy_std_dev
0.01125640617422735
cost_per_task_usd
0.007000730858468677
output_tokens_per_task
17260.92807424594
latency_seconds_per_task
651.373611755607
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
59
benchmark_evaluation
true
openrouter_permaslug
google/gemma-4-31b-it-20260402
source model id
google/gemma-4-31b-it-20260402
Original result ↗

OpenRouter Benchmarks

75.619%

source id
openRouterBenchmarks
leaderboard name
tau2-bench Verified Airline · Accuracy · auto-routing
dimension
agentic
metric
tau2-bench Verified Airline · Accuracy
rank
24
sample size
350
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
7
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
tau_bench_verified_airline
name
tau2-bench Verified Airline
record_kind
aggregate
provider_name
auto-routing
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-31T19:35:25.708Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
tau2-bench Verified Airline · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
350
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/tau2-bench-airline
measurements
accuracy
0.7561904285714286
accuracy_std_dev
0.03184434649468365
cost_per_task_usd
0.03082173142857143
output_tokens_per_task
11747.214285714286
latency_seconds_per_task
402.85049714285714
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
128
benchmark_evaluation
true
openrouter_permaslug
google/gemma-4-31b-it-20260402
source model id
google/gemma-4-31b-it-20260402
Original result ↗

OpenRouter Benchmarks

43.556%

source id
openRouterBenchmarks
leaderboard name
tau2-bench Verified Airline · Accuracy · Parasail
dimension
agentic
metric
tau2-bench Verified Airline · Accuracy
rank
26
sample size
300
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
6
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
tau_bench_verified_airline
name
tau2-bench Verified Airline
record_kind
aggregate
provider_name
Parasail
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-31T19:46:05.351Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
tau2-bench Verified Airline · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
300
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/tau2-bench-airline
measurements
accuracy
0.43555550000000004
accuracy_std_dev
0.03136793204186934
cost_per_task_usd
0.00797013
output_tokens_per_task
4634.386666666666
latency_seconds_per_task
203.44301000000002
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
28
benchmark_evaluation
true
openrouter_permaslug
google/gemma-4-31b-it-20260402
source model id
google/gemma-4-31b-it-20260402
Original result ↗

OpenRouter Benchmarks

82.548%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · DeepInfra
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
27
sample size
3564
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
18
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
gpqa_diamond
name
GPQA Diamond
record_kind
aggregate
provider_name
DeepInfra
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-31T16:45:45.611Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
GPQA Diamond · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
3564
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.8254768888888889
accuracy_std_dev
0.05394208057505227
cost_per_task_usd
0.007830105218855219
output_tokens_per_task
17689.749438832772
latency_seconds_per_task
541.4919023569024
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
71
benchmark_evaluation
true
openrouter_permaslug
google/gemma-4-31b-it-20260402
source model id
google/gemma-4-31b-it-20260402
Original result ↗

OpenRouter Benchmarks

66.667%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · Venice
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
35
sample size
1386
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
7
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
gpqa_diamond
name
GPQA Diamond
record_kind
aggregate
provider_name
Venice
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-31T16:17:15.718Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
GPQA Diamond · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
1386
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.6666665714285714
accuracy_std_dev
0.017285925272620274
cost_per_task_usd
0.00577388023088023
output_tokens_per_task
15830.38961038961
latency_seconds_per_task
421.4543968253968
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
39
benchmark_evaluation
true
openrouter_permaslug
google/gemma-4-31b-it-20260402
source model id
google/gemma-4-31b-it-20260402
Original result ↗

OpenRouter Benchmarks

70.63%

source id
openRouterBenchmarks
leaderboard name
tau2-bench Verified Airline · Accuracy · Novita
dimension
agentic
metric
tau2-bench Verified Airline · Accuracy
rank
36
sample size
349
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
7
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
tau_bench_verified_airline
name
tau2-bench Verified Airline
record_kind
aggregate
provider_name
Novita
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-31T19:57:15.103Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
tau2-bench Verified Airline · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
349
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/tau2-bench-airline
measurements
accuracy
0.7063038567335244
accuracy_std_dev
0.03586910330405172
cost_per_task_usd
0.02153851862464183
output_tokens_per_task
9341.048710601719
latency_seconds_per_task
628.2356618911175
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
52
benchmark_evaluation
true
openrouter_permaslug
google/gemma-4-31b-it-20260402
source model id
google/gemma-4-31b-it-20260402
Original result ↗

Published prices

1–20 / 21

gemma-4-31b-it ↗

$0.16

Inference provider
cheaperInference
Platform
Cheaper Inference
Unit
input_tokens_per_million
Currency
USD
Billing conditions
zdr
false
protocol
chat_completions
service_tier
standard
Billing label
Input tokens (per 1M)

gemma-4-31b-it ↗

$0.49

Inference provider
cheaperInference
Platform
Cheaper Inference
Unit
output_tokens_per_million
Currency
USD
Billing conditions
zdr
false
protocol
chat_completions
service_tier
standard
Billing label
Output tokens (per 1M)

google/gemma-4-31b-it ↗

$0.20

Inference provider
openRouter
Platform
OpenRouter
Unit
input_tokens_per_million
Currency
USD
Billing conditions

google/gemma-4-31b-it ↗

$0.40

Inference provider
openRouter
Platform
OpenRouter
Unit
output_tokens_per_million
Currency
USD
Billing conditions

google/gemma-4-31B-it ↗

$0.25

Inference provider
siliconFlow
Platform
SiliconFlow
Unit
cache_read_tokens_per_million
Currency
USD
Billing conditions
Billing label
Cached input tokens (per 1M)

google/gemma-4-31B-it ↗

$0.75

Inference provider
siliconFlow
Platform
SiliconFlow
Unit
input_tokens_per_million
Currency
USD
Billing conditions
Billing label
Input tokens (per 1M)

google/gemma-4-31B-it ↗

$1.00

Inference provider
siliconFlow
Platform
SiliconFlow
Unit
output_tokens_per_million
Currency
USD
Billing conditions
Billing label
Output tokens (per 1M)

google/gemma-4-31b-it ↗

$0.14

Inference provider
vercelAIGatewayCatalog
Platform
friendli
Unit
input_tokens_per_million
Currency
USD
Billing conditions
endpoint
friendli | google/gemma-4-31b-it
model_type
language
billing_scope
via Vercel AI Gateway
Billing label
Input tokens / 1M
Pricing via
Vercel AI Gateway

google/gemma-4-31b-it ↗

$0.40

Inference provider
vercelAIGatewayCatalog
Platform
friendli
Unit
output_tokens_per_million
Currency
USD
Billing conditions
endpoint
friendli | google/gemma-4-31b-it
model_type
language
billing_scope
via Vercel AI Gateway
Billing label
Output tokens / 1M
Pricing via
Vercel AI Gateway

google/gemma-4-31b-it ↗

$0.10

Inference provider
vercelAIGatewayCatalog
Platform
gmicloud
Unit
cache_read_tokens_per_million
Currency
USD
Billing conditions
endpoint
gmicloud | google/gemma-4-31b-it
model_type
language
quantization
fp8
billing_scope
via Vercel AI Gateway
Billing label
Cached input / 1M
Pricing via
Vercel AI Gateway

google/gemma-4-31b-it ↗

$0.14

Inference provider
vercelAIGatewayCatalog
Platform
gmicloud
Unit
input_tokens_per_million
Currency
USD
Billing conditions
endpoint
gmicloud | google/gemma-4-31b-it
model_type
language
quantization
fp8
billing_scope
via Vercel AI Gateway
Billing label
Input tokens / 1M
Pricing via
Vercel AI Gateway

google/gemma-4-31b-it ↗

$0.40

Inference provider
vercelAIGatewayCatalog
Platform
gmicloud
Unit
output_tokens_per_million
Currency
USD
Billing conditions
endpoint
gmicloud | google/gemma-4-31b-it
model_type
language
quantization
fp8
billing_scope
via Vercel AI Gateway
Billing label
Output tokens / 1M
Pricing via
Vercel AI Gateway

google/gemma-4-31b-it ↗

$0.14

Inference provider
vercelAIGatewayCatalog
Platform
novita
Unit
input_tokens_per_million
Currency
USD
Billing conditions
endpoint
novita | google/gemma-4-31b-it
model_type
language
quantization
bf16
billing_scope
via Vercel AI Gateway
Billing label
Input tokens / 1M
Pricing via
Vercel AI Gateway

google/gemma-4-31b-it ↗

$0.40

Inference provider
vercelAIGatewayCatalog
Platform
novita
Unit
output_tokens_per_million
Currency
USD
Billing conditions
endpoint
novita | google/gemma-4-31b-it
model_type
language
quantization
bf16
billing_scope
via Vercel AI Gateway
Billing label
Output tokens / 1M
Pricing via
Vercel AI Gateway

google/gemma-4-31b-it ↗

$0.06

Inference provider
vercelAIGatewayCatalog
Platform
parasail
Unit
cache_read_tokens_per_million
Currency
USD
Billing conditions
endpoint
parasail | google/gemma-4-31b-it
model_type
language
billing_scope
via Vercel AI Gateway
Billing label
Cached input / 1M
Pricing via
Vercel AI Gateway

google/gemma-4-31b-it ↗

$0.15

Inference provider
vercelAIGatewayCatalog
Platform
parasail
Unit
input_tokens_per_million
Currency
USD
Billing conditions
endpoint
parasail | google/gemma-4-31b-it
model_type
language
billing_scope
via Vercel AI Gateway
Billing label
Input tokens / 1M
Pricing via
Vercel AI Gateway

google/gemma-4-31b-it ↗

$0.40

Inference provider
vercelAIGatewayCatalog
Platform
parasail
Unit
output_tokens_per_million
Currency
USD
Billing conditions
endpoint
parasail | google/gemma-4-31b-it
model_type
language
billing_scope
via Vercel AI Gateway
Billing label
Output tokens / 1M
Pricing via
Vercel AI Gateway

google/gemma-4-31b-it ↗

$0.14

Inference provider
vercelAIGatewayCatalog
Platform
Vercel AI Gateway
Unit
input_tokens_per_million
Currency
USD
Billing conditions
model_type
language
quote_scope
gateway_catalog
billing_scope
Vercel public catalog; endpoint prices may differ
varies_by_provider
true
Billing label
Input tokens / 1M
Pricing via
Vercel AI Gateway

Provider specifications & access

1–14 / 14

google/gemma-4-31B-it ↗

source model id
azure-huggingface/google--gemma-4-31b-it@1
hosting
foundry-managed-compute
platforms
  • Microsoft Foundry
license
apache-2.0
context length
262144
architecture
input_modalities
  • text
  • image
  • video
output_modalities
  • text
capabilities

    google/gemma-4-31B-it ↗

    source model id
    azure-huggingface/google--gemma-4-31b-it@2
    hosting
    foundry-managed-compute
    platforms
    • Microsoft Foundry
    license
    apache-2.0
    context length
    262144
    architecture
    input_modalities
    • text
    • image
    • video
    output_modalities
    • text
    capabilities

      google/gemma-4-31B-it ↗

      source model id
      azure-huggingface/google--gemma-4-31b-it@3
      hosting
      foundry-managed-compute
      platforms
      • Microsoft Foundry
      license
      apache-2.0
      context length
      262144
      architecture
      input_modalities
      • text
      • image
      • video
      output_modalities
      • text
      capabilities

        google/gemma-4-31B-it ↗

        source model id
        azure-huggingface/google--gemma-4-31b-it@4
        hosting
        foundry-managed-compute
        platforms
        • Microsoft Foundry
        license
        apache-2.0
        context length
        262144
        architecture
        input_modalities
        • text
        • image
        • video
        output_modalities
        • text
        capabilities

          google/gemma-4-31B-it ↗

          source model id
          azure-huggingface/google--gemma-4-31b-it@5
          hosting
          foundry-managed-compute
          platforms
          • Microsoft Foundry
          license
          apache-2.0
          context length
          262144
          architecture
          input_modalities
          • text
          • image
          • video
          output_modalities
          • text
          capabilities

            google/gemma-4-31B-it ↗

            source model id
            azure-huggingface/google--gemma-4-31b-it@6
            hosting
            foundry-managed-compute
            platforms
            • Microsoft Foundry
            license
            apache-2.0
            context length
            262144
            architecture
            input_modalities
            • text
            • image
            • video
            output_modalities
            • text
            capabilities

              google/gemma-4-31B-it ↗

              source model id
              azure-huggingface/google--gemma-4-31b-it@7
              hosting
              foundry-managed-compute
              platforms
              • Microsoft Foundry
              license
              apache-2.0
              context length
              262144
              architecture
              input_modalities
              • text
              • image
              • video
              output_modalities
              • text
              capabilities

                gemma-4-31b-it ↗

                source model id
                gemma-4-31b-it
                hosting
                gateway
                platforms
                • Cheaper Inference
                context length
                262144
                max output tokens
                262144
                architecture
                input_modalities
                • text
                • image
                output_modalities
                • text
                capabilities
                • reasoning
                • structuredOutputs
                • toolCalling
                • visionInput

                gemma-4-31b-it ↗

                source model id
                google/gemma-4-31b-it
                hosting
                nvidia-catalog
                platforms
                • NVIDIA Build / NIM
                context length
                262144
                architecture
                input_modalities
                • text
                • image
                • video
                output_modalities
                • text
                capabilities
                • toolCalling
                • reasoning

                Gemma 4 31B IT ↗

                source model id
                google/gemma-4-31b-it
                hosting
                vercel-gateway
                platforms
                • Vercel AI Gateway
                context length
                262144
                max output tokens
                131072
                architecture
                input_modalities
                • text
                • image
                • pdf
                output_modalities
                • text
                capabilities
                • reasoning
                • structuredOutputs
                • toolCalling
                • visionInput

                gemma-4-31B ↗

                source model id
                google/gemma-4-31B-it
                hosting
                siliconflow-hosted
                platforms
                • SiliconFlow
                max output tokens
                262144
                architecture
                input_modalities
                • text
                • image
                output_modalities
                • text
                capabilities
                • promptCaching
                • reasoning
                • structuredOutputs
                • toolCalling

                google/gemma-4-31B-it ↗

                source model id
                HuggingFace/google-gemma-4-31b-it@2
                hosting
                foundry-managed-compute
                platforms
                • Microsoft Foundry
                license
                apache-2.0
                architecture
                input_modalities
                • text
                output_modalities
                • text
                capabilities

                  gemma-4-31B-it ↗

                  source model id
                  publishers/google/models/gemma4@gemma-4-31b-it
                  hosting
                  vertex-self-deployed
                  platforms
                  • Google Cloud Vertex AI
                  architecture
                  input_modalities
                  • text
                  output_modalities
                  • text
                  capabilities

                    google/gemma-4-31B-it ↗

                    source model id
                    publishers/hf-google/models/gemma-4-31b-it@001
                    hosting
                    vertex-self-deployed
                    platforms
                    • Google Cloud Vertex AI
                    architecture
                    input_modalities
                    • image
                    • text
                    output_modalities
                    • text
                    capabilities
                    • multimodal
                    • visionInput

                    Safety assessments

                    1–1 / 1

                    Foundation Model Transparency Index 2025 ↗

                    41 / 100

                    source id
                    stanfordFMTI
                    canonical model id
                    google/gemma-4-31b-it-20260402
                    taxonomy
                    transparency
                    hazard category
                    transparency
                    hazard name
                    Organization transparency
                    policy area
                    Google
                    assessment type
                    organization-level disclosure
                    metric
                    disclosure_score
                    result label
                    41 / 100
                    sample size
                    100
                    updated at
                    2026-11-09T18:08:05.227Z

                    Source specification observations

                    1–16 / 16

                    azureFoundry ↗

                    field
                    context_length
                    value
                    262144
                    normalized value
                    262144
                    scope
                    source-model
                    source id
                    azureFoundry
                    confidence
                    1

                    azureFoundry ↗

                    field
                    context_length
                    value
                    262144
                    normalized value
                    262144
                    scope
                    source-model
                    source id
                    azureFoundry
                    confidence
                    1

                    azureFoundry ↗

                    field
                    context_length
                    value
                    262144
                    normalized value
                    262144
                    scope
                    source-model
                    source id
                    azureFoundry
                    confidence
                    1

                    azureFoundry ↗

                    field
                    context_length
                    value
                    262144
                    normalized value
                    262144
                    scope
                    source-model
                    source id
                    azureFoundry
                    confidence
                    1

                    azureFoundry ↗

                    field
                    context_length
                    value
                    262144
                    normalized value
                    262144
                    scope
                    source-model
                    source id
                    azureFoundry
                    confidence
                    1

                    azureFoundry ↗

                    field
                    context_length
                    value
                    262144
                    normalized value
                    262144
                    scope
                    source-model
                    source id
                    azureFoundry
                    confidence
                    1

                    azureFoundry ↗

                    field
                    context_length
                    value
                    262144
                    normalized value
                    262144
                    scope
                    source-model
                    source id
                    azureFoundry
                    confidence
                    1

                    nvidiaNim ↗

                    field
                    context_length
                    value
                    262144
                    normalized value
                    262144
                    scope
                    source-model
                    source id
                    nvidiaNim
                    confidence
                    1

                    openRouter ↗

                    field
                    context_length
                    value
                    262144
                    normalized value
                    262144
                    scope
                    source-model
                    source id
                    openRouter
                    confidence
                    1

                    openRouter ↗

                    field
                    context_length
                    value
                    262144
                    normalized value
                    262144
                    scope
                    source-model
                    source id
                    openRouter
                    confidence
                    1

                    siliconFlow ↗

                    field
                    context_length
                    value
                    262K
                    scope
                    source-model
                    source id
                    siliconFlow
                    confidence
                    1

                    vercelAIGatewayCatalog ↗

                    field
                    context_length
                    value
                    262144
                    normalized value
                    262144
                    scope
                    source-model
                    source id
                    vercelAIGatewayCatalog
                    confidence
                    1

                    openRouter ↗

                    field
                    max_output_tokens
                    value
                    16384
                    normalized value
                    16384
                    scope
                    source-model
                    source id
                    openRouter
                    confidence
                    1

                    openRouter ↗

                    field
                    max_output_tokens
                    value
                    32768
                    normalized value
                    32768
                    scope
                    source-model
                    source id
                    openRouter
                    confidence
                    1

                    siliconFlow ↗

                    field
                    max_output_tokens
                    value
                    262K
                    scope
                    source-model
                    source id
                    siliconFlow
                    confidence
                    1

                    vercelAIGatewayCatalog ↗

                    field
                    max_output_tokens
                    value
                    131072
                    normalized value
                    131072
                    scope
                    source-model
                    source id
                    vercelAIGatewayCatalog
                    confidence
                    1

                    Original sources

                    1–20 / 32

                    SiliconFlow · Model page ↗

                    Hugging Face · Model repository ↗

                    Google Cloud · Model page ↗

                    Google Cloud · Model page ↗

                    Microsoft Foundry · Model page ↗

                    Microsoft Foundry · Model page ↗

                    Microsoft Foundry · Model page ↗

                    Microsoft Foundry · Model page ↗

                    Microsoft Foundry · Model page ↗

                    Microsoft Foundry · Model page ↗

                    Microsoft Foundry · Model page ↗

                    Microsoft Foundry · Model page ↗

                    NVIDIA · Model page ↗

                    OpenRouter · Model page ↗

                    OpenRouter · Model page ↗

                    SiliconFlow · Playground ↗

                    Vercel AI Gateway · Model page ↗

                    Microsoft Foundry · Model documentation ↗

                    NVIDIA · Documentation ↗

                    SiliconFlow · API documentation ↗

                    Source links

                    1–12 / 12

                    Hugging Face model card

                    1–1 / 1

                    Explore more models

                    Models

                    Loading
                    EVIDENCE / COMPARISON

                    Benchmarks

                    Explore every benchmark

                    Loading

                    These charts contain model-linked results. The charts above also include unlinked source profiles. Compare scores within each evaluation. Model quality, usage, transparency and system performance measure different things.

                    Install TensorTrace