DeepSeek V4 Pro

DeepSeek V4 Pro is a large-scale Mixture-of-Experts model from DeepSeek with 1.6T total parameters and 49B activated parameters, supporting a 1M-token context window. It is designed for advanced reasoning, coding, and long-horizon agent workflows, with strong performance across knowledge, math, and software engineering benchmarks.

Built on the same architecture as DeepSeek V4 Flash, it introduces a hybrid attention system for efficient long-context processing. Reasoning efforts high and xhigh are supported; xhigh maps to max reasoning. It is well suited for complex workloads such as full-codebase analysis, multi-step automation, and large-scale information synthesis, where both capability and efficiency are critical

deepseek/deepseek-v4-pro

Organization: DeepSeek
Family: deepseek-thinking
Providers: 63
Context: 1,000,000
Output limit: 384,000
Knowledge: 2025-05
Release: 2026-04-24
Updated: —
Weights: Open
Input: text
Output: text
Capabilities: —

Providers

Compare every cataloged offer for this model, including its provider, route, token limits, price, and capabilities.

88 offers · USD per million tokens

Provider	Model ID	Context	Output	Input / output · 1M	Reasoning	Tools	Structured	Details
Abacus	deepseek-ai/DeepSeek-V4-Pro	1,000,000	32,768	— / —	—	—	—	View
ai&	deepseek-ai/deepseek-v4-pro	1,048,576	384,000	— / —	—	—	—	View
AIHubMix	deep-deepseek-v4-pro	1,000,000	384,000	— / —	—	—	—	View
AIHubMix	alicloud-deepseek-v4-pro	1,000,000	384,000	— / —	—	—	—	View
AI/ML API	deepseek/deepseek-v4-pro	1,000,000	384,000	— / —	—	—	—	View
Alibaba	deepseek-v4-pro	1,000,000	384,000	— / —	—	—	—	View
Alibaba (China)	deepseek-v4-pro	1,000,000	384,000	— / —	—	—	—	View
Alibaba Token Plan	deepseek-v4-pro	1,000,000	384,000	— / —	—	—	—	View
Alibaba Token Plan (China)	deepseek-v4-pro	1,000,000	384,000	— / —	—	—	—	View
AnyAPI	deepseek/deepseek-v4-pro	1,000,000	384,000	— / —	—	—	—	View
Arcee	deepseek/deepseek-v4-pro	512,000	384,000	— / —	—	—	—	View
Auriko	deepseek-v4-pro	1,000,000	384,000	— / —	—	—	—	View
Azure	deepseek-v4-pro	1,000,000	384,000	— / —	—	—	—	View
BytePlus ModelArk	deepseek-v4-pro-260425	1,000,000	384,000	— / —	—	—	—	View
ClinePass	cline-pass/deepseek-v4-pro	1,000,000	384,000	— / —	—	—	—	View
Cloudflare AI Gateway	deepseek/deepseek-v4-pro	131,072	384,000	— / —	—	—	—	View
CommandCode	deepseek/deepseek-v4-pro	1,000,000	384,000	— / —	—	—	—	View
Cortecs	deepseek-v4-pro	1,000,000	384,000	— / —	—	—	—	View
CrossModel	deepseek/deepseek-v4-pro	1,000,000	384,000	— / —	—	—	—	View
Deep Infra	deepseek-ai/DeepSeek-V4-Pro	1,048,576	16,384	— / —	—	—	—	View
DevPass (LLM Gateway)	deepseek-v4-pro	1,050,000	384,000	— / —	—	—	—	View
DigitalOcean	deepseek-v4-pro	1,048,576	384,000	— / —	—	—	—	View
EBCloud	DeepSeek-V4-Pro	1,000,000	384,000	— / —	—	—	—	View
Eden AI	deepseek/deepseek-v4-pro	1,000,000	384,000	— / —	—	—	—	View
EmpirioLabs AI	deepseek-v4-pro	1,000,000	393,216	— / —	—	—	—	View
FastRouter	deepseek/deepseek-v4-pro	1,000,000	384,000	— / —	—	—	—	View
FrogBot	deepseek-v4-pro	128,000	8,192	— / —	—	—	—	View
GMI Cloud	deepseek-ai/DeepSeek-V4-Pro	1,048,576	384,000	— / —	—	—	—	View
HPC-AI	deepseek/deepseek-v4-pro	1,002,000	128,000	— / —	—	—	—	View
Hugging Face	deepseek-ai/DeepSeek-V4-Pro	1,048,576	393,216	— / —	—	—	—	View
Impossibl	deepseek/deepseek-v4-pro	1,000,000	384,000	— / —	—	—	—	View
Jalapeno Cloud	DeepSeek-V4-Pro	1,048,576	384,000	— / —	—	—	—	View
Kenari	deepseek-v4-pro	1,000,000	384,000	— / —	—	—	—	View
Kilo Gateway	deepseek/deepseek-v4-pro	1,048,576	393,216	— / —	—	—	—	View
LLM Gateway	bytedance/deepseek-v4-pro	1,048,576	393,216	— / —	—	—	—	View
LLM Gateway	alibaba/deepseek-v4-pro	1,000,000	393,216	— / —	—	—	—	View
LLM Gateway	baidu/deepseek-v4-pro	1,048,576	131,072	— / —	—	—	—	View
LLM Gateway	cloudflare/deepseek-v4-pro	1,048,576	393,216	— / —	—	—	—	View
LLM Gateway	deepseek/deepseek-v4-pro	1,050,000	393,216	— / —	—	—	—	View
LLM Gateway	tencent/deepseek-v4-pro	1,000,000	393,216	— / —	—	—	—	View
LLM Gateway	deepinfra/deepseek-v4-pro	1,048,576	64,000	— / —	—	—	—	View
LLM Gateway	together-ai/deepseek-v4-pro	1,048,576	163,840	— / —	—	—	—	View
LLM Gateway	runware/deepseek-v4-pro	1,048,576	384,000	— / —	—	—	—	View
Merge Gateway	deepseek/deepseek-v4-pro	1,000,000	384,000	— / —	—	—	—	View
Moark	DeepSeek-V4-Pro	1,000,000	384,000	— / —	—	—	—	View
Modelis	deepseek-v4-pro	1,000,000	384,000	— / —	—	—	—	View
Model Oracle AI	deepseek-v4-pro	1,000,000	384,000	— / —	—	—	—	View
Novita AI	deepseek/deepseek-v4-pro	1,048,576	393,216	— / —	—	—	—	View
Ofox	deepseek/deepseek-v4-pro	1,000,000	384,000	— / —	—	—	—	View
Ollama Cloud	deepseek-v4-pro	1,048,576	1,048,576	— / —	—	—	—	View
OpenCode	deepseek-v4-pro	1,000,000	384,000	— / —	—	—	—	View
OpenCode Go	deepseek-v4-pro	1,000,000	384,000	— / —	—	—	—	View
Alibabavia OpenRouter	deepseek/deepseek-v4-pro fp8	1,000,000	393,216	— / —	Yes	Yes	Yes	View
AtlasCloudvia OpenRouter	deepseek/deepseek-v4-pro fp4	1,048,576	393,216	$1.68 / $3.38	Yes	Yes	Yes	View
Azurevia OpenRouter	deepseek/deepseek-v4-pro unknown	1,048,576	384,000	$1.91 / $3.83	Yes	Yes	No	View
Baiduvia OpenRouter	deepseek/deepseek-v4-pro fp8	1,048,576	393,216	$1.69 / $3.38	Yes	Yes	Yes	View
Cloudflarevia OpenRouter	deepseek/deepseek-v4-pro unknown	1,048,576	943,718	$1.15 / $2.55	Yes	Yes	Yes	View
DeepInfravia OpenRouter	deepseek/deepseek-v4-pro fp8	1,048,576	16,384	$1.30 / $2.60	Yes	Yes	Yes	View
DigitalOceanvia OpenRouter	deepseek/deepseek-v4-pro unknown	1,048,576	384,000	$1.044 / $2.088	Yes	Yes	No	View
GMICloudvia OpenRouter	deepseek/deepseek-v4-pro fp8	1,048,576	943,718	$0.957 / $1.914	Yes	Yes	No	View
Novitavia OpenRouter	deepseek/deepseek-v4-pro fp8	1,048,576	393,216	$1.60 / $3.20	Yes	Yes	No	View
Parasailvia OpenRouter	deepseek/deepseek-v4-pro fp8	1,048,576	943,718	$0.45 / $3.48	Yes	Yes	Yes	View
Rekavia OpenRouter	deepseek/deepseek-v4-pro unknown	1,048,576	393,216	$1.05 / $10.50	Yes	Yes	Yes	View
Relacevia OpenRouter	deepseek/deepseek-v4-pro fp4	1,048,576	393,216	$0.30 / $4.20	Yes	Yes	No	View
SiliconFlowvia OpenRouter	deepseek/deepseek-v4-pro fp8	1,048,576	393,216	$1.50162 / $3.135	Yes	Yes	No	View
StreamLakevia OpenRouter	deepseek/deepseek-v4-pro fp8	1,024,000	384,000	$0.9483 / $1.8966	Yes	Yes	Yes	View
Venicevia OpenRouter	deepseek/deepseek-v4-pro unknown	1,000,000	32,768	$1.65 / $3.301	Yes	Yes	Yes	View
Opper	deepseek-v4-pro	1,000,000	65,536	— / —	—	—	—	View
OrcaRouter	deepseek/deepseek-v4-pro	1,000,000	384,000	— / —	—	—	—	View
Poe	empiriolabs/deepseek-v4-pro-el	1,000,000	384,000	— / —	—	—	—	View
Prime Inference	deepseek/deepseek-v4-pro	1,048,576	393,216	— / —	—	—	—	View
Requesty	deepseek-v4-pro	1,000,000	131,072	— / —	—	—	—	View
Requesty	deepseek-v4-pro@eu	1,048,576	1,048,576	— / —	—	—	—	View
routing.run	deepseek-v4-pro	1,000,000	64,000	— / —	—	—	—	View
SCNet Token Plan	DeepSeek-V4-Pro	1,000,000	384,000	— / —	—	—	—	View
SiliconFlow	deepseek-ai/DeepSeek-V4-Pro	1,000,000	384,000	— / —	—	—	—	View
SiliconFlow (China)	deepseek-ai/DeepSeek-V4-Pro	1,049,000	393,000	— / —	—	—	—	View
Standard Compute	deepseek/deepseek-v4-pro	1,000,000	50,000	— / —	—	—	—	View
TensorX	deepseek/deepseek-v4-pro	1,048,576	384,000	— / —	—	—	—	View
TokenGo	deepseek/deepseek-v4-pro	1,000,000	384,000	— / —	—	—	—	View
UnoRouter	deepseek-v4-pro:free	1,000,000	384,000	— / —	—	—	—	View
UnoRouter	deepseek-v4-pro	1,000,000	384,000	— / —	—	—	—	View
Venice AI	deepseek-v4-pro	1,000,000	262,144	— / —	—	—	—	View
Vercel AI Gateway	deepseek/deepseek-v4-pro	1,048,600	1,048,600	— / —	—	—	—	View
Vivgrid	deepseek-v4-pro	1,000,000	384,000	— / —	—	—	—	View
Volcengine Ark Coding Plan	deepseek-v4-pro	1,000,000	384,000	— / —	—	—	—	View
Wally	deepseek-v4-pro-fast	1,048,576	65,536	— / —	—	—	—	View
ZenMux	deepseek/deepseek-v4-pro	1,000,000	384,000	— / —	—	—	—	View

Prices and limits apply to each offer. A dash means the value is not provided. OpenRouter offers show the hosting provider and routing channel separately.

Pricing

Compare token rates and cache charges across available routes and hosting configurations.

USD per million tokens

Input from$0.30Per 1M input tokens

Output from$1.8966Per 1M output tokens

Available offers88Across 63 providers

Lowest listed input and output prices across the offers above; they may belong to different providers. These are base rates, before conditional pricing, discounts or additional fees.

Endpoint price comparisonLatest observed snapshot · 10 lowest-priced configurations

InputOutput

Relacerelace/fp4

$0.30$4.20

Parasailparasail/fp8

$0.45$3.48

StreamLakestreamlake/fp8

$0.9483$1.8966

GMICloudgmicloud/fp8

$0.957$1.914

DigitalOceandigitalocean

$1.044$2.088

Rekareka

$1.05$10.50

Cloudflarecloudflare

$1.15$2.55

DeepInfradeepinfra/fp8

$1.30$2.60

SiliconFlowsiliconflow/fp8

$1.50162$3.135

Novitanovita/fp8

$1.60$3.20

deepseek/deepseek-v4-proObserved Oct 10, 2026, 11:31 AM UTC

Input: $0.9483
Output: $1.8966
Cache read: $0.079025

Hosting prices through OpenRouter All endpoints and cache rates

Endpoint	Input / 1M	Output / 1M	Cache read / 1M	Cache write / 1M
Alibabaalibaba/fp8	—	—	—	—
AtlasCloudatlas-cloud/fp4	$1.68	$3.38	$0.13	—
Azureazure/us	$1.91	$3.83	$0.16	—
Baidubaidu/fp8	$1.69	$3.38	$0.14	—
Cloudflarecloudflare	$1.15	$2.55	$0.20	—
DeepInfradeepinfra/fp8	$1.30	$2.60	$0.10	—
DigitalOceandigitalocean	$1.044	$2.088	$0.2088	—
GMICloudgmicloud/fp8	$0.957	$1.914	$0.07975	—
Novitanovita/fp8	$1.60	$3.20	$0.135	—
Parasailparasail/fp8	$0.45	$3.48	$0.10	—
Rekareka	$1.05	$10.50	$0.21	—
Relacerelace/fp4	$0.30	$4.20	$0.23	—
SiliconFlowsiliconflow/fp8	$1.50162	$3.135	$0.135	—
StreamLakestreamlake/fp8	$0.9483	$1.8966	$0.079025	—
Venicevenice	$1.65	$3.301	$0.33	—

Each row is a hosting configuration. Open an offer in Providers for its full pricing conditions and additional fees. Endpoint observations may differ from the routing catalog's base price.

Performance

Review reported latency and output-throughput percentiles from the latest observed rolling window.

Latest 30-minute observation

Time to first token · milliseconds

Latency distributionRolling 30-minute snapshot · 6 endpoints

P50P75P90P99

AtlasCloudatlas-cloud/fp4

1,574–13,354.72 ms

Azureazure/us

2,651–15,105.29 ms

Baidubaidu/fp8

1,083–8,978.08 ms

Cloudflarecloudflare

1,149–21,044.64 ms

DeepInfradeepinfra/fp8

1,112–6,638.29 ms

DigitalOceandigitalocean

1,855.5–32,889.28 ms

All latency percentiles 14 endpoints

Endpoint	P50	P75	P90	P99
AtlasCloudatlas-cloud/fp4	1,574 ms	2,188.5 ms	3,133.5 ms	13,354.72 ms
Azureazure/us	2,651 ms	4,869 ms	7,602.1 ms	15,105.29 ms
Baidubaidu/fp8	1,083 ms	1,513 ms	2,322.2 ms	8,978.08 ms
Cloudflarecloudflare	1,149 ms	1,799 ms	4,213.8 ms	21,044.64 ms
DeepInfradeepinfra/fp8	1,112 ms	1,700 ms	2,526.4 ms	6,638.29 ms
DigitalOceandigitalocean	1,855.5 ms	2,628 ms	4,265 ms	32,889.28 ms
GMICloudgmicloud/fp8	2,458 ms	3,344 ms	5,244.8 ms	81,545.58 ms
Novitanovita/fp8	1,781 ms	2,726 ms	4,331.9 ms	9,569.35 ms
Parasailparasail/fp8	1,690 ms	2,847.5 ms	5,674.8 ms	94,084.56 ms
Rekareka	877 ms	1,301 ms	2,215 ms	43,371.2 ms
Relacerelace/fp4	1,025 ms	1,581 ms	2,302.7 ms	6,420.59 ms
SiliconFlowsiliconflow/fp8	1,938 ms	2,605 ms	3,435.8 ms	7,355.73 ms
StreamLakestreamlake/fp8	1,403.5 ms	2,079.25 ms	3,519 ms	23,642.33 ms
Venicevenice	1,603 ms	2,197 ms	3,066.4 ms	6,545.92 ms

Output throughput · tokens per second

Throughput distributionRolling 30-minute snapshot · 6 endpoints

P50P75P90P99

AtlasCloudatlas-cloud/fp4

44–71.02 t/s

Azureazure/us

75–189 t/s

Baidubaidu/fp8

47–71.07 t/s

Cloudflarecloudflare

46–60 t/s

DeepInfradeepinfra/fp8

42–114 t/s

DigitalOceandigitalocean

18–45.52 t/s

All throughput percentiles 14 endpoints

Endpoint	P50	P75	P90	P99
AtlasCloudatlas-cloud/fp4	44 t/s	51 t/s	58 t/s	71.02 t/s
Azureazure/us	75 t/s	120 t/s	156 t/s	189 t/s
Baidubaidu/fp8	47 t/s	58 t/s	62 t/s	71.07 t/s
Cloudflarecloudflare	46 t/s	52 t/s	55 t/s	60 t/s
DeepInfradeepinfra/fp8	42 t/s	56 t/s	72 t/s	114 t/s
DigitalOceandigitalocean	18 t/s	24 t/s	32 t/s	45.52 t/s
GMICloudgmicloud/fp8	30 t/s	45 t/s	57 t/s	84 t/s
Novitanovita/fp8	55 t/s	70 t/s	83 t/s	118.53 t/s
Parasailparasail/fp8	30 t/s	42 t/s	54 t/s	93.09 t/s
Rekareka	45.5 t/s	53 t/s	62 t/s	79.91 t/s
Relacerelace/fp4	78 t/s	97 t/s	116 t/s	152 t/s
SiliconFlowsiliconflow/fp8	35 t/s	40 t/s	43 t/s	49 t/s
StreamLakestreamlake/fp8	35 t/s	38 t/s	43 t/s	50 t/s
Venicevenice	55 t/s	68 t/s	87 t/s	153 t/s

Percentiles describe the observed request distribution. These are snapshots of a rolling window, not a historical time series.

Uptime

Compare successful request rates across the latest 5-minute, 30-minute, and 24-hour observations.

Successful requests by window

Availability by endpointCurrent overlapping windows · 15 endpoints

Endpoint5 minutes30 minutes24 hours

Alibabaalibaba/fp8Observed Oct 10, 2026, 11:31 AM UTC

100%

—

AtlasCloudatlas-cloud/fp4Observed Oct 10, 2026, 11:31 AM UTC

100%

99.69%

Azureazure/usObserved Oct 10, 2026, 11:31 AM UTC

100%

Baidubaidu/fp8Observed Oct 10, 2026, 11:31 AM UTC

100%

98.57%

CloudflarecloudflareObserved Oct 10, 2026, 11:31 AM UTC

100%

99.58%

DeepInfradeepinfra/fp8Observed Oct 10, 2026, 11:31 AM UTC

100%

DigitalOceandigitaloceanObserved Oct 10, 2026, 11:31 AM UTC

100%

99.93%

GMICloudgmicloud/fp8Observed Oct 10, 2026, 11:31 AM UTC

100%

98.82%

Novitanovita/fp8Observed Oct 10, 2026, 11:31 AM UTC

100%

Parasailparasail/fp8Observed Oct 10, 2026, 11:31 AM UTC

100%

88.89%

92.85%

RekarekaObserved Oct 10, 2026, 11:31 AM UTC

100%

99.48%

Relacerelace/fp4Observed Oct 10, 2026, 11:31 AM UTC

100%

SiliconFlowsiliconflow/fp8Observed Oct 10, 2026, 11:31 AM UTC

40%

10.34%

95.69%

StreamLakestreamlake/fp8Observed Oct 10, 2026, 11:31 AM UTC

100%

99.72%

VeniceveniceObserved Oct 10, 2026, 11:31 AM UTC

100%

96.99%

Reported by OpenRouter at the time of collection; rate-limited requests are excluded. A dash means no measurement was reported. This is not a live availability check.

Benchmarks

Review attributed evaluation results on each source's original scale.

External evaluations

deepseek/deepseek-v4-proObserved Oct 10, 2026, 11:31 AM UTC

Benchmark profileArtificial Analysis · shared source scale from 0 to 60

Coding

59.4

Agentic

26.3

Intelligence

30.4

Design Arena	Category	Elo	Win rate	Rank
agents	fullstack	948	22.1%	43
agents	godotgamedev	1,059	34%	32
agents	webapps	1,000	26.4%	42
models	3d	1,259	56.8%	26
models	asciiart	1,155	46.4%	40
models	codecategories	1,244	51.9%	39
models	dataviz	1,206	48.7%	52
models	gamedev	1,236	52.3%	39
models	svg	1,139	45.4%	48
models	uicomponent	1,225	50.6%	46
models	website	1,239	50.5%	45

Sources: Artificial Analysis and Design Arena, via OpenRouter. Scores retain their original scales and are not OpenAI Suite ratings.

Detailed evaluations

Evaluation scoresAccuracy or item F1 · higher is better

gpqa diamondDeepSeek: DeepSeek V4 Pro 0423

87.278%

tau bench verified airlineDeepSeek: DeepSeek V4 Pro 0423

76.857%

Evaluation	Configuration	Source	Results	Details
models · dataviz	DeepSeek-V4-Prodeepseek/deepseek-v4-pro-20260423	Design Arena	1,200 Elo	Details
models · svg	DeepSeek-V4-Prodeepseek/deepseek-v4-pro-20260423	Design Arena	1,128 Elo	Details
models · codecategories	DeepSeek-V4-Prodeepseek/deepseek-v4-pro-20260423	Design Arena	1,231 Elo	Details
gpqa diamond	DeepSeek: DeepSeek V4 Pro 0423deepseek/deepseek-v4-pro-20260423	OpenRouter	87.278%	Details
models · gamedev	DeepSeek-V4-Prodeepseek/deepseek-v4-pro-20260423	Design Arena	1,221 Elo	Details
models · uicomponent	DeepSeek-V4-Prodeepseek/deepseek-v4-pro-20260423	Design Arena	1,211 Elo	Details
models · asciiart	DeepSeek-V4-Prodeepseek/deepseek-v4-pro-20260423	Design Arena	1,143 Elo	Details
models · website	DeepSeek-V4-Prodeepseek/deepseek-v4-pro-20260423	Design Arena	1,227 Elo	Details
Composite indices	DeepSeek V4 Pro 0813 (High)deepseek/deepseek-v4-pro-20260423	Artificial Analysis	Intelligence 20.8 · Coding — · Agentic —	Details
models · 3d	DeepSeek-V4-Prodeepseek/deepseek-v4-pro-20260423	Design Arena	1,233 Elo	Details
tau bench verified airline	DeepSeek: DeepSeek V4 Pro 0423deepseek/deepseek-v4-pro-20260423	OpenRouter	76.857%	Details

Sources: Artificial Analysis, Design Arena and OpenRouter. Evaluation configurations and scales differ; results are not interchangeable. Updated Nov 14, 2026.

Apps & session costs

Compare observed median session costs for applications using this model, grouped by conversation length.

30-day sample · through Nov 1, 2026

Application	Model route	Turns	Median cost / session	Details
Claude Code	deepseek/deepseek-v4-pro-20260423	2–9 turns	$0.027756	Details
Hermes Agent	deepseek/deepseek-v4-pro-20260423	1 turn	$0.003897	Details
Claude Code	deepseek/deepseek-v4-pro-20260423	10–49 turns	$0.201093	Details
Claude Code	deepseek/deepseek-v4-pro-20260423	50+ turns	$1.335787	Details
Claude Code	deepseek/deepseek-v4-pro-20260423	1 turn	$0.00957	Details
Hermes Agent	deepseek/deepseek-v4-pro-20260423	2–9 turns	$0.026573	Details
Hermes Agent	deepseek/deepseek-v4-pro-20260423	50+ turns	$1.444748	Details
Kilo Code	deepseek/deepseek-v4-pro-20260423	10–49 turns	$0.228141	Details
Hermes Agent	deepseek/deepseek-v4-pro-20260423	10–49 turns	$0.206383	Details
Kilo Code	deepseek/deepseek-v4-pro-20260423	2–9 turns	$0.063541	Details

Source: OpenRouter, as of Nov 9, 2026. Published under CC BY 4.0. These are observed costs per session, not token prices or a forecast for your workload. Apps are compared separately.

Activity

Daily prompt and completion tokens reported for this model's routes in OpenRouter's top 50. Missing days have no published value.

Last 90 complete UTC days

Daily token usage12,605,526,430,426 reported tokens · 90 days with observations

319B0

2026-08-16
298,691,253,867 tokens

2026-08-17
319,451,835,281 tokens

2026-08-18
316,688,850,869 tokens

2026-08-19
269,746,511,450 tokens

2026-08-20
210,459,293,332 tokens

2026-08-21
217,631,001,218 tokens

2026-08-22
246,231,949,121 tokens

2026-08-23
270,469,254,442 tokens

2026-08-24
312,273,659,994 tokens

2026-08-25
310,151,715,462 tokens

2026-08-26
257,753,442,407 tokens

2026-08-27
243,998,048,085 tokens

2026-08-28
233,345,355,232 tokens

2026-08-29
210,832,905,140 tokens

2026-08-30
200,519,709,382 tokens

2026-08-31
218,413,220,489 tokens

2026-09-01
214,528,761,252 tokens

2026-09-02
205,281,925,754 tokens

2026-09-03
206,898,472,775 tokens

2026-09-04
159,532,852,650 tokens

2026-09-05
122,190,238,487 tokens

2026-09-06
126,373,181,897 tokens

2026-09-07
147,374,972,870 tokens

2026-09-08
144,494,119,246 tokens

2026-09-09
165,410,399,480 tokens

2026-09-10
236,380,709,780 tokens

2026-09-11
244,111,965,845 tokens

2026-09-12
188,343,474,587 tokens

2026-09-13
188,395,634,348 tokens

2026-09-14
208,374,709,023 tokens

2026-09-15
211,419,532,056 tokens

2026-09-16
155,865,974,323 tokens

2026-09-17
146,210,750,924 tokens

2026-09-18
139,260,853,910 tokens

2026-09-19
99,978,116,931 tokens

2026-09-20
91,640,202,055 tokens

2026-09-21
117,523,530,302 tokens

2026-09-22
123,764,816,691 tokens

2026-09-23
115,227,912,286 tokens

2026-09-24
119,484,361,664 tokens

2026-09-25
108,537,763,468 tokens

2026-09-26
78,763,803,146 tokens

2026-09-27
82,249,732,328 tokens

2026-09-28
110,406,328,493 tokens

2026-09-29
106,862,904,977 tokens

2026-09-30
102,486,678,479 tokens

2026-10-01
106,932,019,971 tokens

2026-10-02
109,843,127,240 tokens

2026-10-03
96,040,132,183 tokens

2026-10-04
98,615,208,779 tokens

2026-10-05
119,588,153,674 tokens

2026-10-06
118,773,197,307 tokens

2026-10-07
113,664,022,354 tokens

2026-10-08
101,213,950,087 tokens

2026-10-09
95,602,812,143 tokens

2026-10-10
88,832,032,822 tokens

2026-10-11
96,534,624,676 tokens

2026-10-12
102,222,920,951 tokens

2026-10-13
108,652,416,656 tokens

2026-10-14
107,008,034,556 tokens

2026-10-15
96,159,351,469 tokens

2026-10-16
85,513,777,671 tokens

2026-10-17
67,636,629,568 tokens

2026-10-18
86,457,482,874 tokens

2026-10-19
103,207,510,624 tokens

2026-10-20
99,997,626,289 tokens

2026-10-21
94,983,535,345 tokens

2026-10-22
90,952,595,861 tokens

2026-10-23
78,575,177,484 tokens

2026-10-24
80,577,607,257 tokens

2026-10-25
94,373,553,266 tokens

2026-10-26
87,474,329,794 tokens

2026-10-27
88,861,554,352 tokens

2026-10-28
85,281,875,414 tokens

2026-10-29
89,635,675,616 tokens

2026-10-30
83,433,743,514 tokens

2026-10-31
64,357,034,414 tokens

2026-11-01
74,085,512,368 tokens

2026-11-02
103,878,760,400 tokens

2026-11-03
85,325,698,866 tokens

2026-11-04
78,669,410,698 tokens

2026-11-05
81,209,622,343 tokens

2026-11-06
74,388,416,059 tokens

2026-11-07
65,690,325,563 tokens

2026-11-08
80,210,144,012 tokens

2026-11-09
81,261,263,686 tokens

2026-11-10
90,556,547,807 tokens

2026-11-11
79,885,573,825 tokens

2026-11-12
84,834,068,427 tokens

2026-11-13
80,496,686,363 tokens

Aug 16, 2026Nov 13, 2026

Daily observations · 90 days

Date (UTC)	Reported tokens	Reported routes
Nov 13, 2026	80,496,686,363	1
Nov 12, 2026	84,834,068,427	1
Nov 11, 2026	79,885,573,825	1
Nov 10, 2026	90,556,547,807	1
Nov 9, 2026	81,261,263,686	1
Nov 8, 2026	80,210,144,012	1
Nov 7, 2026	65,690,325,563	1
Nov 6, 2026	74,388,416,059	1
Nov 5, 2026	81,209,622,343	1
Nov 4, 2026	78,669,410,698	1
Nov 3, 2026	85,325,698,866	1
Nov 2, 2026	103,878,760,400	1
Nov 1, 2026	74,085,512,368	1
Oct 31, 2026	64,357,034,414	1
Oct 30, 2026	83,433,743,514	1
Oct 29, 2026	89,635,675,616	1
Oct 28, 2026	85,281,875,414	1
Oct 27, 2026	88,861,554,352	1
Oct 26, 2026	87,474,329,794	1
Oct 25, 2026	94,373,553,266	1
Oct 24, 2026	80,577,607,257	1
Oct 23, 2026	78,575,177,484	1
Oct 22, 2026	90,952,595,861	1
Oct 21, 2026	94,983,535,345	1
Oct 20, 2026	99,997,626,289	1
Oct 19, 2026	103,207,510,624	1
Oct 18, 2026	86,457,482,874	1
Oct 17, 2026	67,636,629,568	1
Oct 16, 2026	85,513,777,671	1
Oct 15, 2026	96,159,351,469	1
Oct 14, 2026	107,008,034,556	1
Oct 13, 2026	108,652,416,656	1
Oct 12, 2026	102,222,920,951	1
Oct 11, 2026	96,534,624,676	1
Oct 10, 2026	88,832,032,822	1
Oct 9, 2026	95,602,812,143	1
Oct 8, 2026	101,213,950,087	1
Oct 7, 2026	113,664,022,354	1
Oct 6, 2026	118,773,197,307	1
Oct 5, 2026	119,588,153,674	1
Oct 4, 2026	98,615,208,779	1
Oct 3, 2026	96,040,132,183	1
Oct 2, 2026	109,843,127,240	1
Oct 1, 2026	106,932,019,971	1
Sep 30, 2026	102,486,678,479	1
Sep 29, 2026	106,862,904,977	1
Sep 28, 2026	110,406,328,493	1
Sep 27, 2026	82,249,732,328	1
Sep 26, 2026	78,763,803,146	1
Sep 25, 2026	108,537,763,468	1
Sep 24, 2026	119,484,361,664	1
Sep 23, 2026	115,227,912,286	1
Sep 22, 2026	123,764,816,691	1
Sep 21, 2026	117,523,530,302	1
Sep 20, 2026	91,640,202,055	1
Sep 19, 2026	99,978,116,931	1
Sep 18, 2026	139,260,853,910	1
Sep 17, 2026	146,210,750,924	1
Sep 16, 2026	155,865,974,323	1
Sep 15, 2026	211,419,532,056	1
Sep 14, 2026	208,374,709,023	1
Sep 13, 2026	188,395,634,348	1
Sep 12, 2026	188,343,474,587	1
Sep 11, 2026	244,111,965,845	1
Sep 10, 2026	236,380,709,780	1
Sep 9, 2026	165,410,399,480	1
Sep 8, 2026	144,494,119,246	1
Sep 7, 2026	147,374,972,870	1
Sep 6, 2026	126,373,181,897	1
Sep 5, 2026	122,190,238,487	1
Sep 4, 2026	159,532,852,650	1
Sep 3, 2026	206,898,472,775	1
Sep 2, 2026	205,281,925,754	1
Sep 1, 2026	214,528,761,252	1
Aug 31, 2026	218,413,220,489	1
Aug 30, 2026	200,519,709,382	1
Aug 29, 2026	210,832,905,140	1
Aug 28, 2026	233,345,355,232	1
Aug 27, 2026	243,998,048,085	1
Aug 26, 2026	257,753,442,407	1
Aug 25, 2026	310,151,715,462	1
Aug 24, 2026	312,273,659,994	1
Aug 23, 2026	270,469,254,442	1
Aug 22, 2026	246,231,949,121	1
Aug 21, 2026	217,631,001,218	1
Aug 20, 2026	210,459,293,332	1
Aug 19, 2026	269,746,511,450	1
Aug 18, 2026	316,688,850,869	1
Aug 17, 2026	319,451,835,281	1
Aug 16, 2026	298,691,253,867	1

Source: OpenRouter (openrouter.ai/rankings), as of Nov 14, 2026. CC BY 4.0. Totals include only individually published routes. Tokenizers differ by provider; the aggregated “other” category is never assigned to a model.

Technical details & API

Explore specifications, supported parameters, reasoning controls, and API identifiers for each OpenRouter route.

Current route specifications

deepseek/deepseek-v4-proObserved Oct 10, 2026, 11:31 AM UTC

Context: 1,048,576
Maximum output: 384,000
Tokenizer: DeepSeek
Input: text
Output: text
Moderation: No
Added to OpenRouter: 2026-04-24

Base URLhttps://openrouter.ai/api/v1

Model IDdeepseek/deepseek-v4-pro

Version IDdeepseek/deepseek-v4-pro-20260423

Weights IDdeepseek-ai/DeepSeek-V4-Pro

Reasoning

Required: No
Enabled by default: Not reported
Default effort: high
Supported efforts: xhigh, high

Supported parameters

frequency_penalty
include_reasoning
logit_bias
logprobs
max_completion_tokens
max_tokens
min_p
presence_penalty
reasoning
reasoning_effort
repetition_penalty
response_format
seed
stop
structured_outputs
temperature
tool_choice
tools
top_k
top_logprobs
top_p

Default parameters

top p: 1
temperature: 1

Frequently asked questions

Answers based on the model metadata and serving offers currently in the catalog.

How much does DeepSeek V4 Pro cost?

Listed rates start at $0.30 per million input tokens and $1.8966 per million output tokens. Prices vary by provider and configuration; see Pricing for details.

Which providers offer DeepSeek V4 Pro?

There are 88 cataloged offers across 63 providers. Compare model IDs, prices and limits in Providers.

What is the context window?

The cataloged model context is 1,000,000 tokens. Each serving endpoint may apply a different limit.

Does it support tools and structured output?

Tool calling: Not reported. Structured output: Not reported. Support can vary by endpoint.

Resources

Hugging Face ↗

Catalog history · 3 recorded revisions

Metadata observations since this model was first cataloged. These are not model release versions.

Oct 20, 2026 · b213c28db852
Sep 14, 2026 · 30e90fb0153d
Sep 12, 2026 · cb27a78cd069