DeepSeek: DeepSeek V4.1 Flash 1017

DeepSeek V4.1 Flash 1017 is the October 17, 2026 snapshot of DeepSeek V4.1 Flash, served by DeepSeek under the `deepseek-flash` API model name. V4.1 Flash is a sparse mixture-of-experts model from De…

deepseek/deepseek-v4.1-flash-20261017

DeepSeek V4.1 Flash 1017 is the October 17, 2026 snapshot of DeepSeek V4.1 Flash, served by DeepSeek under the deepseek-flash API model name. V4.1 Flash is a sparse mixture-of-experts model from DeepSeek, and the first built on the company's Causal Encoder-Decoder (CED) architecture. It activates 8B parameters on input and 16B on output from a 552B-parameter backbone, an asymmetric split that keeps per-token compute low relative to the model's total size. Image understanding is native to the architecture, with visual and text embeddings trained jointly from the start of pre-training rather than added afterward as in the earlier experimental V4 Flash Vision Exp.

It is suited for coding, terminal, and computer-use agents, along with long-horizon tasks that must run to completion across many steps and long-context analysis. Compressed KV caching cuts cache memory to roughly a quarter of the previous Flash generation, significantly reducing costs on agentic workloads. DeepSeek positions it as the cost-efficient tier of the V4.1 family and reports that it exceeds V4 Pro on performance, speed, and task completion time.

Specifications & capabilities

Input
text, image
Output
text
Context window · tokens
1048576
Published model ID
DeepSeek-V4.1-Flash-1017
Capabilities
multimodal, promptCaching, reasoning, structuredOutputs, toolCalling, visionInput
Description source ↗

Evaluations

1–20 / 83

OpenRouter Benchmarks

92.812%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · Tenstorrent
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
1
sample size
160
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
Tenstorrent
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-11-04T18:00:19.969Z
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
160
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.9281247875
accuracy_std_dev
0.003304499999999988
cost_per_task_usd
0.01232809375
output_tokens_per_task
10178.75625
latency_seconds_per_task
251.38681875
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
deepseek/deepseek-v4.1-flash-20261017
source model id
deepseek/deepseek-v4.1-flash-20261017
Original result ↗

OpenRouter Benchmarks

92.34%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · Near AI
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
1
sample size
396
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
Near AI
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-31T10:06:11.748Z
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
396
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.9234005000000001
accuracy_std_dev
0.005892500000000023
cost_per_task_usd
0.02620280303030303
output_tokens_per_task
21737.315656565657
latency_seconds_per_task
592.1412247474748
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
4
benchmark_evaluation
true
openrouter_permaslug
deepseek/deepseek-v4.1-flash-20261017
source model id
deepseek/deepseek-v4.1-flash-20261017
Original result ↗

OpenRouter Benchmarks

91.863%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · SiliconFlow
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
1
sample size
594
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
SiliconFlow
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-31T08:50:50.713Z
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
594
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.9186303333333333
accuracy_std_dev
0.010408091158111372
cost_per_task_usd
0.025292042087542085
output_tokens_per_task
20957.713804713803
latency_seconds_per_task
160.4023451178451
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
deepseek/deepseek-v4.1-flash-20261017
source model id
deepseek/deepseek-v4.1-flash-20261017
Original result ↗

OpenRouter Benchmarks

91.583%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · DekaLLM
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
1
sample size
198
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
DekaLLM
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-11-04T09:00:09.993Z
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
198
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.915825
cost_per_task_usd
0.014663378787878788
output_tokens_per_task
21419.76767676768
latency_seconds_per_task
96.92635858585858
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
12
benchmark_evaluation
true
openrouter_permaslug
deepseek/deepseek-v4.1-flash-20261017
source model id
deepseek/deepseek-v4.1-flash-20261017
Original result ↗

OpenRouter Benchmarks

91.497%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · CoreWeave
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
1
sample size
196
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
CoreWeave
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-11-04T18:00:08.219Z
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
196
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.9149660000000001
cost_per_task_usd
0.022535153061224493
output_tokens_per_task
18639.61224489796
latency_seconds_per_task
106.1645357142857
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
deepseek/deepseek-v4.1-flash-20261017
source model id
deepseek/deepseek-v4.1-flash-20261017
Original result ↗

OpenRouter Benchmarks

91.145%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · GMICloud
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
1
sample size
990
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
gpqa_diamond
name
GPQA Diamond
record_kind
aggregate
provider_name
GMICloud
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-11-10T06:01:40.563Z
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
990
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.9114477999999999
accuracy_std_dev
0.0039265801609033885
cost_per_task_usd
0.018160035353535352
output_tokens_per_task
22106.812121212122
latency_seconds_per_task
129.7026080808081
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
26
benchmark_evaluation
true
openrouter_permaslug
deepseek/deepseek-v4.1-flash-20261017
source model id
deepseek/deepseek-v4.1-flash-20261017
Original result ↗

OpenRouter Benchmarks

90.909%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · StreamLake
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
1
sample size
396
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
StreamLake
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-31T08:36:23.829Z
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
396
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.9090910000000001
accuracy_std_dev
0.003367000000000009
cost_per_task_usd
0.018701613636363638
output_tokens_per_task
22059.484848484848
latency_seconds_per_task
137.00363383838382
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
30
benchmark_evaluation
true
openrouter_permaslug
deepseek/deepseek-v4.1-flash-20261017
source model id
deepseek/deepseek-v4.1-flash-20261017
Original result ↗

OpenRouter Benchmarks

90.909%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · StepFun
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
1
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
StepFun
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-11-02T17:24:31.225Z
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
0.9090908333333333
accuracy_std_dev
0.007076201038614481
cost_per_task_usd
0.01709530134680135
output_tokens_per_task
21725.516835016835
latency_seconds_per_task
84.45594696969698
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
1
benchmark_evaluation
true
openrouter_permaslug
deepseek/deepseek-v4.1-flash-20261017
source model id
deepseek/deepseek-v4.1-flash-20261017
Original result ↗

OpenRouter Benchmarks

90.741%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · Cohere
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
1
sample size
396
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
Cohere
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-31T08:13:11.036Z
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
396
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.9074074999999998
accuracy_std_dev
0.008417500000000022
cost_per_task_usd
0.014508287878787877
output_tokens_per_task
21790.61111111111
latency_seconds_per_task
90.08695707070707
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
7
benchmark_evaluation
true
openrouter_permaslug
deepseek/deepseek-v4.1-flash-20261017
source model id
deepseek/deepseek-v4.1-flash-20261017
Original result ↗

OpenRouter Benchmarks

90.432%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · Mistral
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
1
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
Mistral
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-11-11T02:59:16.591Z
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
0.904321
accuracy_std_dev
0.006024476685433623
cost_per_task_usd
0.025462516835016838
output_tokens_per_task
21063.152356902356
latency_seconds_per_task
81.18970454545455
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
5
benchmark_evaluation
true
openrouter_permaslug
deepseek/deepseek-v4.1-flash-20261017
source model id
deepseek/deepseek-v4.1-flash-20261017
Original result ↗

OpenRouter Benchmarks

90.32%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · Particle AI
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
1
sample size
792
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
gpqa_diamond
name
GPQA Diamond
record_kind
aggregate
provider_name
Particle AI
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-11-03T01:49:06.293Z
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
792
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.9031985
accuracy_std_dev
0.0075759444460740165
cost_per_task_usd
0.01504592297979798
output_tokens_per_task
17203.179292929293
latency_seconds_per_task
86.73782702020202
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
2
benchmark_evaluation
true
openrouter_permaslug
deepseek/deepseek-v4.1-flash-20261017
source model id
deepseek/deepseek-v4.1-flash-20261017
Original result ↗

OpenRouter Benchmarks

89.798%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · Ionstream
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
1
sample size
990
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
gpqa_diamond
name
GPQA Diamond
record_kind
aggregate
provider_name
Ionstream
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-11-10T16:47:27.230Z
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
990
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.8979796
accuracy_std_dev
0.008689115757083655
cost_per_task_usd
0.014924614141414142
output_tokens_per_task
23160.81515151515
latency_seconds_per_task
134.09519090909092
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
5
benchmark_evaluation
true
openrouter_permaslug
deepseek/deepseek-v4.1-flash-20261017
source model id
deepseek/deepseek-v4.1-flash-20261017
Original result ↗

OpenRouter Benchmarks

89.169%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · OpenInference
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
1
sample size
594
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
OpenInference
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-31T09:01:06.311Z
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
594
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.8916946666666665
accuracy_std_dev
0.009944030347678707
cost_per_task_usd
0.011122356902356903
output_tokens_per_task
25444.905723905726
latency_seconds_per_task
302.42450673400674
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
7
benchmark_evaluation
true
openrouter_permaslug
deepseek/deepseek-v4.1-flash-20261017
source model id
deepseek/deepseek-v4.1-flash-20261017
Original result ↗

OpenRouter Benchmarks

77.778%

source id
openRouterBenchmarks
leaderboard name
tau2-bench Verified Airline · Accuracy · AtlasCloud
dimension
agentic
metric
tau2-bench Verified Airline · Accuracy
rank
1
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
AtlasCloud
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-31T11:43:38.949Z
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
0.777778
accuracy_std_dev
0.01257094435593444
cost_per_task_usd
0.016507473333333335
output_tokens_per_task
10861.266666666666
latency_seconds_per_task
92.51076666666667
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
29
benchmark_evaluation
true
openrouter_permaslug
deepseek/deepseek-v4.1-flash-20261017
source model id
deepseek/deepseek-v4.1-flash-20261017
Original result ↗

OpenRouter Benchmarks

72.556%

source id
openRouterBenchmarks
leaderboard name
tau2-bench Verified Airline · Accuracy · StepFun
dimension
agentic
metric
tau2-bench Verified Airline · Accuracy
rank
1
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
StepFun
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-11-02T17:45:53.936Z
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
0.7255556666666666
accuracy_std_dev
0.016960557004087128
cost_per_task_usd
0.012908883333333333
output_tokens_per_task
10836.613333333333
latency_seconds_per_task
53.54427
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
1
benchmark_evaluation
true
openrouter_permaslug
deepseek/deepseek-v4.1-flash-20261017
source model id
deepseek/deepseek-v4.1-flash-20261017
Original result ↗

OpenRouter Benchmarks

92.281%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · DeepInfra
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
2
sample size
570
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
DeepInfra
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-31T08:40:34.158Z
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
570
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.9228069789473685
accuracy_std_dev
0.011126928806578514
cost_per_task_usd
0.00580749649122807
output_tokens_per_task
19521.97543859649
latency_seconds_per_task
218.61469122807017
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
deepseek/deepseek-v4.1-flash-20261017
source model id
deepseek/deepseek-v4.1-flash-20261017
Original result ↗

OpenRouter Benchmarks

92.088%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · DigitalOcean
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
2
sample size
594
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
DigitalOcean
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-31T08:09:30.885Z
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
594
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.9208753333333334
accuracy_std_dev
0.003636621356271357
cost_per_task_usd
0.019656516835016836
output_tokens_per_task
20830.57239057239
latency_seconds_per_task
90.4379696969697
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
deepseek/deepseek-v4.1-flash-20261017
source model id
deepseek/deepseek-v4.1-flash-20261017
Original result ↗

OpenRouter Benchmarks

91.758%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · PrimaLabs
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
2
sample size
182
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
PrimaLabs
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-11-06T15:33:21.027Z
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
182
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.9175825384615385
accuracy_std_dev
0.009135000000000004
cost_per_task_usd
0.03656871978021978
output_tokens_per_task
10074.60989010989
latency_seconds_per_task
44.23142857142857
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
3
benchmark_evaluation
true
openrouter_permaslug
deepseek/deepseek-v4.1-flash-20261017
source model id
deepseek/deepseek-v4.1-flash-20261017
Original result ↗

OpenRouter Benchmarks

91.311%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · Baidu
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
2
sample size
750
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
gpqa_diamond
name
GPQA Diamond
record_kind
aggregate
provider_name
Baidu
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-11-10T18:06:00.757Z
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
750
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.91311096
accuracy_std_dev
0.009099497963486788
cost_per_task_usd
0.01717951733333333
output_tokens_per_task
20403.964
latency_seconds_per_task
130.3304
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
12
benchmark_evaluation
true
openrouter_permaslug
deepseek/deepseek-v4.1-flash-20261017
source model id
deepseek/deepseek-v4.1-flash-20261017
Original result ↗

OpenRouter Benchmarks

90.544%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · Makora
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
2
sample size
987
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
gpqa_diamond
name
GPQA Diamond
record_kind
aggregate
provider_name
Makora
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-31T08:31:11.848Z
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
987
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.9054373465045593
accuracy_std_dev
0.0037554479359991244
cost_per_task_usd
0.026207787234042552
output_tokens_per_task
21741.29989868288
latency_seconds_per_task
127.14136068895644
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
12
benchmark_evaluation
true
openrouter_permaslug
deepseek/deepseek-v4.1-flash-20261017
source model id
deepseek/deepseek-v4.1-flash-20261017
Original result ↗

Published prices

61–80 / 110

deepseek/deepseek-v4.1-flash-1017 ↗

$0.08

Inference provider
vercelAIGatewayCatalog
Platform
pareto
Unit
input_tokens_per_million
Currency
USD
Billing conditions
endpoint
pareto | deepseek/deepseek-v4.1-flash-1017
model_type
language
peak_pricing
windows
  • days_of_week
    • 1
    • 2
    • 3
    • 4
    • 5
    end_minute_utc
    240
    start_minute_utc
    60
  • days_of_week
    • 1
    • 2
    • 3
    • 4
    • 5
    end_minute_utc
    600
    start_minute_utc
    360
multiplier
2
billing_scope
via Vercel AI Gateway
Billing label
Input tokens / 1M
Pricing via
Vercel AI Gateway

deepseek/deepseek-v4.1-flash-1017 ↗

$0.30

Inference provider
vercelAIGatewayCatalog
Platform
pareto
Unit
output_tokens_per_million
Currency
USD
Billing conditions
endpoint
pareto | deepseek/deepseek-v4.1-flash-1017
model_type
language
peak_pricing
windows
  • days_of_week
    • 1
    • 2
    • 3
    • 4
    • 5
    end_minute_utc
    240
    start_minute_utc
    60
  • days_of_week
    • 1
    • 2
    • 3
    • 4
    • 5
    end_minute_utc
    600
    start_minute_utc
    360
multiplier
2
billing_scope
via Vercel AI Gateway
Billing label
Output tokens / 1M
Pricing via
Vercel AI Gateway

deepseek/deepseek-v4.1-flash-1017 ↗

$0.02

Inference provider
vercelAIGatewayCatalog
Platform
particle
Unit
cache_read_tokens_per_million
Currency
USD
Billing conditions
endpoint
particle | deepseek/deepseek-v4.1-flash-1017
model_type
language
peak_pricing
windows
  • end_minute_utc
    30
    start_minute_utc
    480
multiplier
2
billing_scope
via Vercel AI Gateway
Billing label
Cached input / 1M
Pricing via
Vercel AI Gateway

deepseek/deepseek-v4.1-flash-1017 ↗

$0.02

Inference provider
vercelAIGatewayCatalog
Platform
particle
Unit
cache_read_tokens_per_million
Currency
USD
Billing conditions
region
scope
zone
geo_region
us
endpoint
particle | deepseek/deepseek-v4.1-flash-1017
model_type
language
billing_scope
via Vercel AI Gateway
Billing label
Cached input / 1M
Pricing via
Vercel AI Gateway

deepseek/deepseek-v4.1-flash-1017 ↗

$0.13

Inference provider
vercelAIGatewayCatalog
Platform
particle
Unit
input_tokens_per_million
Currency
USD
Billing conditions
endpoint
particle | deepseek/deepseek-v4.1-flash-1017
model_type
language
peak_pricing
windows
  • end_minute_utc
    30
    start_minute_utc
    480
multiplier
2
billing_scope
via Vercel AI Gateway
Billing label
Input tokens / 1M
Pricing via
Vercel AI Gateway

deepseek/deepseek-v4.1-flash-1017 ↗

$0.13

Inference provider
vercelAIGatewayCatalog
Platform
particle
Unit
input_tokens_per_million
Currency
USD
Billing conditions
region
scope
zone
geo_region
us
endpoint
particle | deepseek/deepseek-v4.1-flash-1017
model_type
language
billing_scope
via Vercel AI Gateway
Billing label
Input tokens / 1M
Pricing via
Vercel AI Gateway

deepseek/deepseek-v4.1-flash-1017 ↗

$0.50

Inference provider
vercelAIGatewayCatalog
Platform
particle
Unit
output_tokens_per_million
Currency
USD
Billing conditions
endpoint
particle | deepseek/deepseek-v4.1-flash-1017
model_type
language
peak_pricing
windows
  • end_minute_utc
    30
    start_minute_utc
    480
multiplier
2
billing_scope
via Vercel AI Gateway
Billing label
Output tokens / 1M
Pricing via
Vercel AI Gateway

deepseek/deepseek-v4.1-flash-1017 ↗

$0.50

Inference provider
vercelAIGatewayCatalog
Platform
particle
Unit
output_tokens_per_million
Currency
USD
Billing conditions
region
scope
zone
geo_region
us
endpoint
particle | deepseek/deepseek-v4.1-flash-1017
model_type
language
billing_scope
via Vercel AI Gateway
Billing label
Output tokens / 1M
Pricing via
Vercel AI Gateway

deepseek/deepseek-v4.1-flash-1017 ↗

$0.0030

Inference provider
vercelAIGatewayCatalog
Platform
relace
Unit
cache_read_tokens_per_million
Currency
USD
Billing conditions
endpoint
relace | deepseek/deepseek-v4.1-flash-1017
model_type
language
billing_scope
via Vercel AI Gateway
Billing label
Cached input / 1M
Pricing via
Vercel AI Gateway

deepseek/deepseek-v4.1-flash-1017 ↗

$0.04

Inference provider
vercelAIGatewayCatalog
Platform
relace
Unit
input_tokens_per_million
Currency
USD
Billing conditions
endpoint
relace | deepseek/deepseek-v4.1-flash-1017
model_type
language
billing_scope
via Vercel AI Gateway
Billing label
Input tokens / 1M
Pricing via
Vercel AI Gateway

deepseek/deepseek-v4.1-flash-1017 ↗

$0.50

Inference provider
vercelAIGatewayCatalog
Platform
relace
Unit
output_tokens_per_million
Currency
USD
Billing conditions
endpoint
relace | deepseek/deepseek-v4.1-flash-1017
model_type
language
billing_scope
via Vercel AI Gateway
Billing label
Output tokens / 1M
Pricing via
Vercel AI Gateway

deepseek/deepseek-v4.1-flash-1017 ↗

$0.01

Inference provider
vercelAIGatewayCatalog
Platform
runinfra
Unit
cache_read_tokens_per_million
Currency
USD
Billing conditions
endpoint
runinfra | deepseek/deepseek-v4.1-flash-1017
model_type
language
billing_scope
via Vercel AI Gateway
Billing label
Cached input / 1M
Pricing via
Vercel AI Gateway

deepseek/deepseek-v4.1-flash-1017 ↗

$0.12

Inference provider
vercelAIGatewayCatalog
Platform
runinfra
Unit
input_tokens_per_million
Currency
USD
Billing conditions
endpoint
runinfra | deepseek/deepseek-v4.1-flash-1017
model_type
language
billing_scope
via Vercel AI Gateway
Billing label
Input tokens / 1M
Pricing via
Vercel AI Gateway

deepseek/deepseek-v4.1-flash-1017 ↗

$0.48

Inference provider
vercelAIGatewayCatalog
Platform
runinfra
Unit
output_tokens_per_million
Currency
USD
Billing conditions
endpoint
runinfra | deepseek/deepseek-v4.1-flash-1017
model_type
language
billing_scope
via Vercel AI Gateway
Billing label
Output tokens / 1M
Pricing via
Vercel AI Gateway

deepseek/deepseek-v4.1-flash-1017 ↗

$0.01

Inference provider
vercelAIGatewayCatalog
Platform
runware
Unit
cache_read_tokens_per_million
Currency
USD
Billing conditions
endpoint
runware | deepseek/deepseek-v4.1-flash-1017
model_type
language
billing_scope
via Vercel AI Gateway
Billing label
Cached input / 1M
Pricing via
Vercel AI Gateway

deepseek/deepseek-v4.1-flash-1017 ↗

$0.15

Inference provider
vercelAIGatewayCatalog
Platform
runware
Unit
input_tokens_per_million
Currency
USD
Billing conditions
endpoint
runware | deepseek/deepseek-v4.1-flash-1017
model_type
language
billing_scope
via Vercel AI Gateway
Billing label
Input tokens / 1M
Pricing via
Vercel AI Gateway

deepseek/deepseek-v4.1-flash-1017 ↗

$0.60

Inference provider
vercelAIGatewayCatalog
Platform
runware
Unit
output_tokens_per_million
Currency
USD
Billing conditions
endpoint
runware | deepseek/deepseek-v4.1-flash-1017
model_type
language
billing_scope
via Vercel AI Gateway
Billing label
Output tokens / 1M
Pricing via
Vercel AI Gateway

deepseek/deepseek-v4.1-flash-1017 ↗

$0.01

Inference provider
vercelAIGatewayCatalog
Platform
sail-research
Unit
cache_read_tokens_per_million
Currency
USD
Billing conditions
endpoint
sail-research | deepseek/deepseek-v4.1-flash-1017
model_type
language
billing_scope
via Vercel AI Gateway
Billing label
Cached input / 1M
Pricing via
Vercel AI Gateway

deepseek/deepseek-v4.1-flash-1017 ↗

$0.09

Inference provider
vercelAIGatewayCatalog
Platform
sail-research
Unit
input_tokens_per_million
Currency
USD
Billing conditions
endpoint
sail-research | deepseek/deepseek-v4.1-flash-1017
model_type
language
billing_scope
via Vercel AI Gateway
Billing label
Input tokens / 1M
Pricing via
Vercel AI Gateway

deepseek/deepseek-v4.1-flash-1017 ↗

$0.30

Inference provider
vercelAIGatewayCatalog
Platform
sail-research
Unit
output_tokens_per_million
Currency
USD
Billing conditions
endpoint
sail-research | deepseek/deepseek-v4.1-flash-1017
model_type
language
billing_scope
via Vercel AI Gateway
Billing label
Output tokens / 1M
Pricing via
Vercel AI Gateway

Provider specifications & access

1–5 / 5

deepseek-ai/DeepSeek-V4.1-Flash-1017 ↗

source model id
azure-huggingface/deepseek-ai--deepseek-v4.1-flash-1017@1
hosting
foundry-managed-compute
platforms
  • Microsoft Foundry
license
MIT
context length
1048576
architecture
input_modalities
  • text
  • image
output_modalities
  • text
capabilities

    deepseek-ai/DeepSeek-V4.1-Flash-1017 ↗

    source model id
    azure-huggingface/deepseek-ai--deepseek-v4.1-flash-1017@2
    hosting
    foundry-managed-compute
    platforms
    • Microsoft Foundry
    license
    MIT
    context length
    1048576
    architecture
    input_modalities
    • text
    • image
    output_modalities
    • text
    capabilities

      deepseek-ai/DeepSeek-V4.1-Flash-1017 ↗

      source model id
      azure-huggingface/deepseek-ai--deepseek-v4.1-flash-1017@3
      hosting
      foundry-managed-compute
      platforms
      • Microsoft Foundry
      license
      MIT
      context length
      1048576
      architecture
      input_modalities
      • text
      • image
      output_modalities
      • text
      capabilities

        deepseek-v4.1-flash-1017 ↗

        source model id
        @cf/deepseek-ai/deepseek-v4.1-flash-1017
        hosting
        cloudflare-hosted
        platforms
        • Cloudflare Workers AI
        context length
        1048576
        architecture
        input_modalities
        • text
        • image
        output_modalities
        • text
        capabilities
        • toolCalling

        DeepSeek V4.1 Flash 1017 ↗

        source model id
        deepseek/deepseek-v4.1-flash-1017
        hosting
        vercel-gateway
        platforms
        • Vercel AI Gateway
        context length
        1048576
        max output tokens
        262144
        architecture
        input_modalities
        • text
        • image
        output_modalities
        • text
        capabilities
        • promptCaching
        • reasoning
        • structuredOutputs
        • toolCalling
        • visionInput
        regions
        • eu
        • us

        Safety assessments

        1–1 / 1

        Foundation Model Transparency Index 2025 ↗

        32 / 100

        source id
        stanfordFMTI
        canonical model id
        deepseek/deepseek-v4.1-flash-20261017
        taxonomy
        transparency
        hazard category
        transparency
        hazard name
        Organization transparency
        policy area
        DeepSeek
        assessment type
        organization-level disclosure
        metric
        disclosure_score
        result label
        32 / 100
        sample size
        100
        updated at
        2026-11-09T18:08:05.227Z

        Source specification observations

        1–11 / 11

        azureFoundry ↗

        field
        context_length
        value
        1048576
        normalized value
        1048576
        scope
        source-model
        source id
        azureFoundry
        confidence
        1

        azureFoundry ↗

        field
        context_length
        value
        1048576
        normalized value
        1048576
        scope
        source-model
        source id
        azureFoundry
        confidence
        1

        azureFoundry ↗

        field
        context_length
        value
        1048576
        normalized value
        1048576
        scope
        source-model
        source id
        azureFoundry
        confidence
        1

        openRouter ↗

        field
        context_length
        value
        1048576
        normalized value
        1048576
        scope
        source-model
        source id
        openRouter
        confidence
        1

        openRouter ↗

        field
        context_length
        value
        1048576
        normalized value
        1048576
        scope
        source-model
        source id
        openRouter
        confidence
        1

        openRouter ↗

        field
        context_length
        value
        1048576
        normalized value
        1048576
        scope
        source-model
        source id
        openRouter
        confidence
        1

        vercelAIGatewayCatalog ↗

        field
        context_length
        value
        1048576
        normalized value
        1048576
        scope
        source-model
        source id
        vercelAIGatewayCatalog
        confidence
        1

        openRouter ↗

        field
        max_output_tokens
        value
        943718
        normalized value
        943718
        scope
        source-model
        source id
        openRouter
        confidence
        1

        openRouter ↗

        field
        max_output_tokens
        value
        393216
        normalized value
        393216
        scope
        source-model
        source id
        openRouter
        confidence
        1

        openRouter ↗

        field
        max_output_tokens
        value
        943718
        normalized value
        943718
        scope
        source-model
        source id
        openRouter
        confidence
        1

        vercelAIGatewayCatalog ↗

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

        Original sources

        1–17 / 17

        Hugging Face · deepseek-ai/DeepSeek-V4.1-Flash-1017 ↗

        Cloudflare AI · Model page ↗

        Microsoft Foundry · Model page ↗

        Microsoft Foundry · Model page ↗

        Microsoft Foundry · Model page ↗

        OpenRouter · Model page ↗

        OpenRouter · Model page ↗

        Vercel AI Gateway · Model page ↗

        Microsoft Foundry · Model documentation ↗

        GitHub · Source snapshot ↗

        Cloudflare AI · Catalog ↗

        GitHub · Catalog ↗

        Hugging Face · Catalog ↗

        Microsoft Foundry · Catalog ↗

        OpenRouter · Catalog ↗

        OpenRouter · Catalog ↗

        Vercel AI Gateway · Catalog ↗

        Source links

        1–9 / 9

        Related links

        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