Qwen: Qwen3.5-9B

Qwen3.5-9B is a multimodal foundation model from the Qwen3.5 family, designed to deliver strong reasoning, coding, and visual understanding in an efficient 9B-parameter architecture. It uses a unifie…

qwen/qwen3.5-9b-20260310

Qwen3.5-9B is a multimodal foundation model from the Qwen3.5 family, designed to deliver strong reasoning, coding, and visual understanding in an efficient 9B-parameter architecture. It uses a unified vision-language design with early fusion of multimodal tokens, allowing the model to process and reason across text and images within the same context.

Specifications & capabilities

Input
text, image, video
Output
text
Context window · tokens
262144
Published model ID
Qwen3.5-9B
Capabilities
multimodal, promptCaching, reasoning, structuredOutputs, toolCalling, visionInput
Description source ↗

Evaluations

1–18 / 18

OpenRouter Benchmarks

74.405%

source id
openRouterBenchmarks
leaderboard name
tau2-bench Verified Airline · Accuracy · Darkbloom
dimension
agentic
metric
tau2-bench Verified Airline · Accuracy
rank
4
sample size
196
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
4
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
Darkbloom
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-11-14T19:18:15.980Z
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
196
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/tau2-bench-airline
measurements
accuracy
0.7440478163265306
accuracy_std_dev
0.024480750157011157
cost_per_task_usd
0.02607208673469388
output_tokens_per_task
54210.87755102041
latency_seconds_per_task
1123.1503724489796
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
8
benchmark_evaluation
true
openrouter_permaslug
qwen/qwen3.5-9b-20260310
source model id
qwen/qwen3.5-9b-20260310
Original result ↗

OpenRouter Benchmarks

75.273%

source id
openRouterBenchmarks
leaderboard name
tau2-bench Verified Airline · Accuracy · Together
dimension
agentic
metric
tau2-bench Verified Airline · Accuracy
rank
5
sample size
397
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
8
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
Together
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-24T19:38:23.433Z
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
397
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/tau2-bench-airline
measurements
accuracy
0.7527289420654912
accuracy_std_dev
0.024253123484087077
cost_per_task_usd
0.06776491435768262
output_tokens_per_task
77465.15365239295
latency_seconds_per_task
1043.1933853904281
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
qwen/qwen3.5-9b-20260310
source model id
qwen/qwen3.5-9b-20260310
Original result ↗

OpenRouter Benchmarks

69.588%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · Darkbloom
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
7
sample size
1980
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
10
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
gpqa_diamond
name
GPQA Diamond
record_kind
aggregate
provider_name
Darkbloom
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-11-14T18:57:17.561Z
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
1980
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.6958755
accuracy_std_dev
0.06752091374714948
cost_per_task_usd
0.007408071717171717
output_tokens_per_task
57767.52626262626
latency_seconds_per_task
1271.1492494949496
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
qwen/qwen3.5-9b-20260310
source model id
qwen/qwen3.5-9b-20260310
Original result ↗

OpenRouter Benchmarks

80.966%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · PrimeIntellect
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
11
sample size
521
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
PrimeIntellect
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-31T16:04:23.357Z
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
521
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.8096608426103648
accuracy_std_dev
0.012817931849127972
cost_per_task_usd
0.0079749443378119
output_tokens_per_task
31380.54894433781
latency_seconds_per_task
289.7114932821497
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
qwen/qwen3.5-9b-20260310
source model id
qwen/qwen3.5-9b-20260310
Original result ↗

OpenRouter Benchmarks

66.667%

source id
openRouterBenchmarks
leaderboard name
tau2-bench Verified Airline · Accuracy · PrimeIntellect
dimension
agentic
metric
tau2-bench Verified Airline · Accuracy
rank
11
sample size
150
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
PrimeIntellect
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-31T19:43:40.322Z
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
150
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/tau2-bench-airline
measurements
accuracy
0.6666666666666666
accuracy_std_dev
0.0524933858267454
cost_per_task_usd
0.03491572666666667
output_tokens_per_task
23860.97333333333
latency_seconds_per_task
274.07564666666667
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
21
benchmark_evaluation
true
openrouter_permaslug
qwen/qwen3.5-9b-20260310
source model id
qwen/qwen3.5-9b-20260310
Original result ↗

OpenRouter Benchmarks

76.19%

source id
openRouterBenchmarks
leaderboard name
tau2-bench Verified Airline · Accuracy · DeepInfra
dimension
agentic
metric
tau2-bench Verified Airline · Accuracy
rank
13
sample size
245
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
DeepInfra
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-09T04:05:56.641Z
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
245
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/tau2-bench-airline
measurements
accuracy
0.7619046734693877
accuracy_std_dev
0.00894270029465374
cost_per_task_usd
0.03367711020408164
output_tokens_per_task
51764.93877551021
latency_seconds_per_task
2596.3409346938774
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
qwen/qwen3.5-9b-20260310
source model id
qwen/qwen3.5-9b-20260310
Original result ↗

OpenRouter Benchmarks

65.649%

source id
openRouterBenchmarks
leaderboard name
tau2-bench Verified Airline · Accuracy · SiliconFlow
dimension
agentic
metric
tau2-bench Verified Airline · Accuracy
rank
22
sample size
344
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
SiliconFlow
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-17T21:28:32.796Z
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
344
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/tau2-bench-airline
measurements
accuracy
0.656492308139535
accuracy_std_dev
0.13396399619194066
cost_per_task_usd
0.02718404941860465
output_tokens_per_task
35003.26453488372
latency_seconds_per_task
1177.0430639534884
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
qwen/qwen3.5-9b-20260310
source model id
qwen/qwen3.5-9b-20260310
Original result ↗

OpenRouter Benchmarks

80.479%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · Parasail
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
23
sample size
1267
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
9
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
2026-10-31T16:04:29.125Z
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
1267
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.8047882407261246
accuracy_std_dev
0.03134006389099038
cost_per_task_usd
0.007205067876874505
output_tokens_per_task
28563.743488555643
latency_seconds_per_task
327.8137805840568
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
qwen/qwen3.5-9b-20260310
source model id
qwen/qwen3.5-9b-20260310
Original result ↗

OpenRouter Benchmarks

79.108%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · Together
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
23
sample size
2051
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
11
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
2026-10-31T16:15:26.662Z
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
2051
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.791077650901999
accuracy_std_dev
0.011802764782558223
cost_per_task_usd
0.00883669673330083
output_tokens_per_task
34825.446123842026
latency_seconds_per_task
380.01700048756703
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
qwen/qwen3.5-9b-20260310
source model id
qwen/qwen3.5-9b-20260310
Original result ↗

OpenRouter Benchmarks

79.424%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · Venice
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
26
sample size
1782
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
9
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
2026-10-17T16:03:34.733Z
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
1782
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.7942385555555556
accuracy_std_dev
0.012314380393228746
cost_per_task_usd
0.004576078563411897
output_tokens_per_task
29957.45061728395
latency_seconds_per_task
256.06055387205384
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
qwen/qwen3.5-9b-20260310
source model id
qwen/qwen3.5-9b-20260310
Original result ↗

OpenRouter Benchmarks

73.536%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · SiliconFlow
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
26
sample size
1383
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
SiliconFlow
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-17T16:03:21.011Z
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
1383
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.7353578611713666
accuracy_std_dev
0.08764063669791727
cost_per_task_usd
0.004913945770065076
output_tokens_per_task
32251.255242227042
latency_seconds_per_task
616.099819956616
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
qwen/qwen3.5-9b-20260310
source model id
qwen/qwen3.5-9b-20260310
Original result ↗

OpenRouter Benchmarks

79.097%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · DeepInfra
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
31
sample size
2042
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
11
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
2026-10-31T16:08:08.537Z
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
2042
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.7909730112634672
accuracy_std_dev
0.015262253874258334
cost_per_task_usd
0.004672809010773752
output_tokens_per_task
30652.18511263467
latency_seconds_per_task
1542.6738849167484
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
qwen/qwen3.5-9b-20260310
source model id
qwen/qwen3.5-9b-20260310
Original result ↗

OpenRouter Benchmarks

65.333%

source id
openRouterBenchmarks
leaderboard name
tau2-bench Verified Airline · Accuracy · Venice
dimension
agentic
metric
tau2-bench Verified Airline · Accuracy
rank
31
sample size
400
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
8
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
Venice
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-17T19:25:29.604Z
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
400
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/tau2-bench-airline
measurements
accuracy
0.653333375
accuracy_std_dev
0.07817377366472963
cost_per_task_usd
0.0266300325
output_tokens_per_task
25080.435
latency_seconds_per_task
292.6225475
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
38
benchmark_evaluation
true
openrouter_permaslug
qwen/qwen3.5-9b-20260310
source model id
qwen/qwen3.5-9b-20260310
Original result ↗

OpenRouter Benchmarks

73.367%

source id
openRouterBenchmarks
leaderboard name
tau2-bench Verified Airline · Accuracy · auto-routing
dimension
agentic
metric
tau2-bench Verified Airline · Accuracy
rank
45
sample size
398
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
8
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
2026-10-31T20:03:31.183Z
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
398
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/tau2-bench-airline
measurements
accuracy
0.7336684371859297
accuracy_std_dev
0.0449684893006689
cost_per_task_usd
0.033184525125628146
output_tokens_per_task
52852.572864321606
latency_seconds_per_task
1207.8773844221107
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
qwen/qwen3.5-9b-20260310
source model id
qwen/qwen3.5-9b-20260310
Original result ↗

OpenRouter Benchmarks

77.395%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · auto-routing
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
86
sample size
2178
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
11
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
gpqa_diamond
name
GPQA Diamond
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
2026-10-31T16:03:21.447Z
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
2178
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.7739515454545453
accuracy_std_dev
0.019647060545183753
cost_per_task_usd
0.005980178604224059
output_tokens_per_task
37567.65151515151
latency_seconds_per_task
705.4164118457301
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
132
benchmark_evaluation
true
openrouter_permaslug
qwen/qwen3.5-9b-20260310
source model id
qwen/qwen3.5-9b-20260310
Original result ↗

Published prices

1–5 / 5

qwen/qwen3.5-9b ↗

$0.05

Inference provider
openRouter
Platform
OpenRouter
Unit
cache_read_tokens_per_million
Currency
USD
Billing conditions

qwen/qwen3.5-9b ↗

$0.05

Inference provider
openRouter
Platform
OpenRouter
Unit
input_tokens_per_million
Currency
USD
Billing conditions

qwen/qwen3.5-9b ↗

$0.08

Inference provider
openRouter
Platform
OpenRouter
Unit
output_tokens_per_million
Currency
USD
Billing conditions

Qwen/Qwen3.5-9B ↗

$0.10

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

Qwen/Qwen3.5-9B ↗

$0.15

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

Provider specifications & access

1–10 / 10

Qwen/Qwen3.5-9B ↗

source model id
azure-huggingface/qwen--qwen3.5-9b@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

    Qwen/Qwen3.5-9B ↗

    source model id
    azure-huggingface/qwen--qwen3.5-9b@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

      Qwen/Qwen3.5-9B ↗

      source model id
      azure-huggingface/qwen--qwen3.5-9b@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

        Qwen/Qwen3.5-9B ↗

        source model id
        azure-huggingface/qwen--qwen3.5-9b@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

          Qwen/Qwen3.5-9B ↗

          source model id
          azure-huggingface/qwen--qwen3.5-9b@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

            Qwen/Qwen3.5-9B ↗

            source model id
            azure-huggingface/qwen--qwen3.5-9b@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

              Qwen/Qwen3.5-9B ↗

              source model id
              HuggingFace/qwen-qwen3.5-9b@1
              hosting
              foundry-managed-compute
              platforms
              • Microsoft Foundry
              license
              apache-2.0
              architecture
              input_modalities
              • text
              output_modalities
              • text
              capabilities

                Qwen/Qwen3.5-9B ↗

                source model id
                publishers/hf-qwen/models/qwen3.5-9b@001
                hosting
                vertex-self-deployed
                platforms
                • Google Cloud Vertex AI
                architecture
                input_modalities
                • image
                • text
                output_modalities
                • text
                capabilities
                • multimodal
                • visionInput

                Qwen3.5-9B ↗

                source model id
                publishers/qwen/models/qwen3-5@qwen3.5-9b
                hosting
                vertex-self-deployed
                platforms
                • Google Cloud Vertex AI
                architecture
                input_modalities
                • text
                output_modalities
                • text
                capabilities

                  Qwen3.5-9B ↗

                  source model id
                  Qwen/Qwen3.5-9B
                  hosting
                  siliconflow-hosted
                  platforms
                  • SiliconFlow
                  max output tokens
                  262144
                  architecture
                  input_modalities
                  • text
                  • image
                  output_modalities
                  • text
                  capabilities
                  • reasoning
                  • structuredOutputs
                  • toolCalling

                  Safety assessments

                  1–1 / 1

                  Foundation Model Transparency Index 2025 ↗

                  26 / 100

                  source id
                  stanfordFMTI
                  canonical model id
                  qwen/qwen3.5-9b-20260310
                  taxonomy
                  transparency
                  hazard category
                  transparency
                  hazard name
                  Organization transparency
                  policy area
                  Alibaba
                  assessment type
                  organization-level disclosure
                  metric
                  disclosure_score
                  result label
                  26 / 100
                  sample size
                  100
                  updated at
                  2026-11-09T18:08:05.227Z

                  Source specification observations

                  1–10 / 10

                  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

                  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

                  openRouter ↗

                  field
                  max_output_tokens
                  value
                  40960
                  normalized value
                  40960
                  scope
                  source-model
                  source id
                  openRouter
                  confidence
                  1

                  siliconFlow ↗

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

                  Original sources

                  1–20 / 25

                  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 ↗

                  OpenRouter · Model page ↗

                  SiliconFlow · Playground ↗

                  Google Cloud · Model documentation ↗

                  Microsoft Foundry · Model documentation ↗

                  SiliconFlow · API documentation ↗

                  GitHub · Model reference ↗

                  Epoch AI · Catalog ↗

                  GitHub · Catalog ↗

                  Google Cloud · Catalog ↗

                  Source links

                  1–9 / 9

                  Hugging Face model card

                  1–1 / 1

                  Qwen/Qwen3.5-9B ↗

                  downloads
                  6029812
                  likes
                  2376
                  pipeline tag
                  image-text-to-text
                  library name
                  transformers
                  License
                  apache-2.0

                  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