DeepSeek model product
DeepSeek-V4-Pro
DeepSeek-V4-Pro-0813 is DeepSeek’s GA production release of DeepSeek-V4-Pro for agentic coding and tool-use workloads.
Updated Oct 23, 2026. Default version: DeepSeek-V4-Pro-0813
LLMBoard Score83.2DeepSeek-V4-Pro-0813
Coverage80%11 benchmark families
Context window1MTokens
Official input price$0.66DeepSeek API
DeepSeek-V4-Pro Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
DeepSeek-V4-Pro Benchmark Results
Benchmark scores for DeepSeek-V4-Pro-0813.
Sort by
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkDSBench-FullStack | Score71.10% | Rank01 | Participants2 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkDSBench-Hard | Score67.20% | Rank01 | Participants3 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkHumanity's Last Exam (with tools, text-only) | Score60.00% | Rank02 | Participants10 | Percentile88.89% | EvidenceC | Evaluated |
| BenchmarkNL2Repo | Score61.50% | Rank03 | Participants25 | Percentile91.67% | EvidenceC | Evaluated |
| BenchmarkToolathlon | Score74.10% | Rank03 | Participants42 | Percentile95.12% | EvidenceC | Evaluated |
| BenchmarkDeepSWE | Score62.70% | Rank06 | Participants13 | Percentile58.33% | EvidenceC | Evaluated |
| BenchmarkHumanity's Last Exam | Score60.00% | Rank07 | Participants109 | Percentile94.44% | EvidenceC | Evaluated |
| BenchmarkLiveBench math (2026-06-25) | Score95.09 score | Rank07 | Participants43 | Percentile85.71% | EvidenceB | EvaluatedN/A |
| BenchmarkCyberGym | Score83.30% | Rank08 | Participants23 | Percentile68.18% | EvidenceC | Evaluated |
| BenchmarkTerminal-Bench 2.1 | Score87.90% | Rank08 | Participants48 | Percentile85.11% | EvidenceC | Evaluated |
| BenchmarkHumanity's Last Exam (no tools, text-only) | Score42.70% | Rank09 | Participants14 | Percentile38.46% | EvidenceC | Evaluated |
| BenchmarkAutomationBench | Score31.80% | Rank15 | Participants27 | Percentile46.15% | EvidenceC | Evaluated |
| BenchmarkLM Arena Agent Bash Recovery Steps | Score4.40% | Rank16 | Participants44 | Percentile65.12% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Tool Hallucination | Score0.55% | Rank16 | Participants44 | Percentile65.12% | EvidenceA | Evaluated |
| BenchmarkAA GPQA Diamond | Score92.83% | Rank16 | Participants199 | Percentile92.42% | EvidenceB | EvaluatedN/A |
| BenchmarkAA CritPt | Score18.00% | Rank18 | Participants200 | Percentile91.46% | EvidenceB | EvaluatedN/A |
| BenchmarkLiveBench reasoning (2026-06-25) | Score85.84 score | Rank18 | Participants43 | Percentile59.52% | EvidenceB | EvaluatedN/A |
| BenchmarkAA Omniscience Accuracy | Score49.10% | Rank19 | Participants201 | Percentile91.00% | EvidenceB | EvaluatedN/A |
| BenchmarkLM Arena Agent Praise Complaint | Score1.02% | Rank20 | Participants44 | Percentile55.81% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Bash Reliability | Score2.20% | Rank21 | Participants44 | Percentile53.49% | EvidenceA | Evaluated |
| BenchmarkAgents' Last Exam | Score25.70% | Rank22 | Participants24 | Percentile8.70% | EvidenceC | Evaluated |
| BenchmarkLM Arena Agent Steerability | Score0.00% | Rank22 | Participants44 | Percentile51.16% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Bash Reliability No Prereq | Score2.36% | Rank22 | Participants45 | Percentile52.27% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Leaderboard | Score0.07% | Rank23 | Participants44 | Percentile48.84% | EvidenceA | Evaluated |
| BenchmarkAA LCR v1.1 | Score80.33% | Rank24 | Participants193 | Percentile88.02% | EvidenceB | EvaluatedN/A |
| BenchmarkLM Arena Agent Task Outcome Explicit | Score-2.36% | Rank25 | Participants44 | Percentile44.19% | EvidenceA | Evaluated |
| BenchmarkLM Arena Webdev | Score1,582.38 rating | Rank25 | Participants107 | Percentile77.36% | EvidenceA | Evaluated |
| BenchmarkAA HLE Text No Tools | Score41.01% | Rank26 | Participants200 | Percentile87.44% | EvidenceB | EvaluatedN/A |
| BenchmarkLiveBench instruction (2026-06-25) | Score67.70 score | Rank26 | Participants43 | Percentile40.48% | EvidenceB | EvaluatedN/A |
| BenchmarkAA SciCode Subtasks | Score51.04% | Rank29 | Participants89 | Percentile68.18% | EvidenceB | EvaluatedN/A |
DeepSeek-V4-Pro Arena Results
Preference and agent-evaluation results for the default version.
Sort by
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenatext factuality | Categoryjapanese | Rank09 | Rating / score1,461.06 | Votes537 | ObservationsN/A | Result date |
| Arenatext style control | Categorykorean | Rank11 | Rating / score1,466.27 | Votes197 | ObservationsN/A | Result date |
| Arenatext factuality | Categorygerman | Rank13 | Rating / score1,441.12 | Votes558 | ObservationsN/A | Result date |
| Arenaagent bash recovery steps | Categoryoverall | Rank16 | Rating / score0.04 | VotesN/A | Observations23.4K | Result date |
| Arenaagent tool hallucination | Categoryoverall | Rank16 | Rating / score0.01 | VotesN/A | Observations765.7K | Result date |
| Arenatext | Categorykorean | Rank18 | Rating / score1,448.57 | Votes197 | ObservationsN/A | Result date |
| Arenatext factuality | Categorykorean | Rank18 | Rating / score1,418.70 | Votes675 | ObservationsN/A | Result date |
| Arenaagent praise complaint | Categoryoverall | Rank20 | Rating / score0.01 | VotesN/A | Observations1.5K | Result date |
| Arenatext factuality | Categoryfrench | Rank21 | Rating / score1,491.12 | Votes1,859 | ObservationsN/A | Result date |
| Arenaagent bash reliability | Categoryoverall | Rank21 | Rating / score0.02 | VotesN/A | Observations81.4K | Result date |
| Arenaagent steerability | Categoryoverall | Rank22 | Rating / score0.00 | VotesN/A | Observations26.9K | Result date |
| Arenaagent bash reliability no prereq | Categoryoverall | Rank22 | Rating / score0.02 | VotesN/A | Observations79.6K | Result date |
| Arenaagent | Categoryoverall | Rank23 | Rating / score0.00 | VotesN/A | Observations878.7K | Result date |
| Arenawebdev | Categorywebdev-react | Rank23 | Rating / score1,588.10 | Votes3,099 | ObservationsN/A | Result date |
| Arenatext | Categorypolish | Rank24 | Rating / score1,460.81 | Votes1,125 | ObservationsN/A | Result date |
| Arenawebdev | Categorywebdev-html | Rank24 | Rating / score1,565.64 | Votes548 | ObservationsN/A | Result date |
| Arenaagent task outcome explicit | Categoryoverall | Rank25 | Rating / score-0.02 | VotesN/A | Observations3.2K | Result date |
| Arenawebdev | Categoryoverall | Rank25 | Rating / score1,582.38 | Votes4,714 | ObservationsN/A | Result date |
| Arenawebdev | Categorywebdev | Rank25 | Rating / score1,582.38 | Votes4,714 | ObservationsN/A | Result date |
| Arenatext | Categoryjapanese | Rank26 | Rating / score1,446.32 | Votes721 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryspanish | Rank26 | Rating / score1,458.31 | Votes1,484 | ObservationsN/A | Result date |
| Arenatext style control | Categoryjapanese | Rank26 | Rating / score1,456.28 | Votes721 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryindustry medicine and healthcare | Rank28 | Rating / score1,489.12 | Votes3,329 | ObservationsN/A | Result date |
| Arenatext | Categorycreative writing | Rank29 | Rating / score1,446.24 | Votes9,724 | ObservationsN/A | Result date |
| Arenatext factuality | Categorymulti turn | Rank29 | Rating / score1,476.09 | Votes9,015 | ObservationsN/A | Result date |
| Arenatext | Categoryfrench | Rank30 | Rating / score1,473.31 | Votes2,210 | ObservationsN/A | Result date |
| Arenatext | Categorymulti turn | Rank30 | Rating / score1,467.62 | Votes10,355 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryindustry legal and government | Rank30 | Rating / score1,478.32 | Votes3,804 | ObservationsN/A | Result date |
| Arenatext factuality | Categorycreative writing | Rank31 | Rating / score1,445.23 | Votes8,610 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry legal and government | Rank31 | Rating / score1,480.36 | Votes825 | ObservationsN/A | Result date |
DeepSeek-V4-Pro Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- $0.66 input, $1.98 output per 1M
- Official provider
- DeepSeek
- Lowest third-party
- From $0.1452 input, $0.4356 output per 1M via NanoGPT
- Tracked offerings
- 42
Sort by
Provider | Provider model ID | Region | Input / 1M | Output / 1M | Context | Updated |
|---|
| ProviderAlibaba Token Plan | Provider model IDdeepseek-v4-pro-0813 | Regionglobal | Input / 1MN/A | Output / 1MN/A | Context1M | Updated |
| ProviderSCNet Token Plan | Provider model IDDeepSeek-V4-Pro-0813 | Regionglobal | Input / 1MN/A | Output / 1MN/A | Context1M | Updated |
| ProviderAlibaba Token Plan (China) | Provider model IDdeepseek-v4-pro-0813 | Regionglobal | Input / 1MN/A | Output / 1MN/A | Context1M | Updated |
| ProviderSenseNova (China) | Provider model IDdeepseek-v4-pro | Regionglobal | Input / 1MN/A | Output / 1MN/A | Context1M | Updated |
| ProviderNanoGPT | Provider model IDdeepseek/deepseek-v4-pro-0813 | Regionglobal | Input / 1M$0.1452 | Output / 1M$0.4356 | Context1M | Updated |
| ProviderRunInfra | Provider model IDdeepseek-ai/DeepSeek-V4-Pro-0813 | Regionglobal | Input / 1M$0.60 | Output / 1M$1.9 | Context1M | Updated |
| ProviderOrcaRouter | Provider model IDdeepseek/deepseek-v4-pro-0813 | Regionglobal | Input / 1M$0.66 | Output / 1M$1.98 | Context1M | Updated |
| ProviderDeepSeek | Provider model IDdeepseek-v4-pro | Regionglobal | Input / 1M$0.66 | Output / 1M$1.98 | Context1M | Updated |
| ProviderOpenRouter | Provider model IDdeepseek/deepseek-v4-pro-0813 | Regionglobal | Input / 1M$0.66 | Output / 1M$1.98 | Context1M | Updated |
| ProviderVercel AI Gateway | Provider model IDdeepseek/deepseek-v4-pro-0813 | Regionglobal | Input / 1M$0.66 | Output / 1M$1.98 | Context1M | Updated |
| ProviderOllama Cloud | Provider model IDdeepseek-v4-pro:0813 | Regionglobal | Input / 1M$0.66 | Output / 1M$1.98 | Context1M | Updated |
| ProviderTempr Gateway | Provider model IDdeepseek/deepseek-v4-pro | Regionglobal | Input / 1M$0.66 | Output / 1M$1.98 | Context1M | Updated |
| ProviderAIHubMix | Provider model IDdeepseek-v4-pro-0813 | Regionglobal | Input / 1M$0.6918 | Output / 1M$2.08 | Context1M | Updated |
| Providerabove.dev | Provider model IDdeepseek-v4-pro | Regionglobal | Input / 1M$0.726 | Output / 1M$2.18 | Context1M | Updated |
| ProviderIteraCompute | Provider model IDdeepseek/deepseek-v4-pro-0813 | Regionglobal | Input / 1M$0.79 | Output / 1M$2.38 | Context1M | Updated |
| ProviderBoundless | Provider model IDdeepseek-v4-pro-0813 | Regionglobal | Input / 1M$1.04 | Output / 1M$2.08 | Context1M | Updated |
| ProviderStandard Compute | Provider model IDdeepseek/deepseek-v4-pro-0813 | Regionglobal | Input / 1M$1.12 | Output / 1M$3.37 | Context1M | Updated |
| ProviderDeep Infra | Provider model IDdeepseek-ai/DeepSeek-V4-Pro-0813 | Regionglobal | Input / 1M$1.3 | Output / 1M$2.6 | Context1M | Updated |
| ProviderCoreWeave | Provider model IDdeepseek-ai/DeepSeek-V4-Pro-0813 | Regionglobal | Input / 1M$1.31 | Output / 1M$3.96 | Context1M | Updated |
| ProviderCloudflare Workers AI | Provider model ID@cf/deepseek-ai/deepseek-v4-pro-0813 | Regionglobal | Input / 1M$1.32 | Output / 1M$3.96 | Context1M | Updated |
| ProviderHugging Face | Provider model IDdeepseek-ai/DeepSeek-V4-Pro-0813 | Regionglobal | Input / 1M$1.32 | Output / 1M$3.96 | Context1M | Updated |
| ProviderDigitalOcean | Provider model IDdeepseek-v4-pro-0813 | Regionglobal | Input / 1M$1.32 | Output / 1M$3.96 | Context1M | Updated |
| ProviderArcee | Provider model IDdeepseek/deepseek-v4-pro-0813 | Regionglobal | Input / 1M$1.32 | Output / 1M$3.96 | Context1M | Updated |
| ProviderTogether AI | Provider model IDdeepseek-ai/DeepSeek-V4-Pro-0813 | Regionglobal | Input / 1M$1.32 | Output / 1M$3.96 | Context1M | Updated |
| ProviderOfox | Provider model IDdeepseek/deepseek-v4-pro-0813 | Regionglobal | Input / 1M$1.32 | Output / 1M$3.96 | Context1M | Updated |
| ProviderRequesty | Provider model IDdeepseek-v4-pro-0813 | Regionglobal | Input / 1M$1.32 | Output / 1M$3.96 | Context1M | Updated |
| ProviderBaseten | Provider model IDdeepseek-ai/DeepSeek-V4-Pro-0813 | Regionglobal | Input / 1M$1.32 | Output / 1M$3.96 | Context1M | Updated |
| ProviderEmpirioLabs AI | Provider model IDdeepseek-v4-pro-0813 | Regionglobal | Input / 1M$1.32 | Output / 1M$3.96 | Context1M | Updated |
| ProviderKilo Gateway | Provider model IDdeepseek/deepseek-v4-pro-0813 | Regionglobal | Input / 1M$1.32 | Output / 1M$3.96 | Context1M | Updated |
| ProviderNebius Token Factory | Provider model IDdeepseek-ai/DeepSeek-V4-Pro-0813 | Regionglobal | Input / 1M$1.32 | Output / 1M$3.96 | Context979K | Updated |
DeepSeek-V4-Pro Runtime Performance
Provider-specific output speed and catalog latency for DeepSeek-V4-Pro-0813. Runtime does not affect the capability score.
No runtime data
No provider-specific speed or latency record is linked to the default version yet.
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
DeepSeek-V4-Pro Specifications
Technical details for the model's default version.
- Version
- DeepSeek-V4-Pro-0813
- Released
- Aug 13, 2026
- Knowledge cutoff
- Unknown
- Parameters
- 1.6T
- Context window
- 1M
- Max output
- 1M
- Inputs
- text
- Outputs
- text
- Open weights
- Yes
- License
- MIT
DeepSeek-V4-Pro Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
Sort by
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionDeepSeek-V4-Pro-0813 | Released | LLMBoard83.16 | Parameters1.6T | Context1M | Max output1M | Open weightsYes | LicenseMIT |
| VersionDeepSeek-V4-Pro-Max | Released | LLMBoardN/A | Parameters1.6T | Context1M | Max output1M | Open weightsYes | LicenseMIT |
Models similar to DeepSeek-V4-Pro
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
DE
DeepSeek-V4.1
DeepSeek
DE
DeepSeek-V4-Flash
DeepSeek
DE
DeepSeek-V3.2
DeepSeek
DE
DeepSeek-R1
DeepSeek
DE
DeepSeek-V3.1
DeepSeek
DE
DeepSeek-V3
DeepSeek
What is DeepSeek-V4-Pro?
Key information about DeepSeek-V4-Pro and its available data.
6T-parameter MoE large language model with 49B activated parameters, low, high, and max reasoning effort, a 1M-token context window, and up to 384K output tokens.
It is served through the OpenAI-compatible Chat Completions and Responses APIs under the ID `deepseek-v4-pro`, with open weights available at `deepseek-ai/DeepSeek-V4-Pro-0813` under the MIT license.
Data as of 2026-10-22.
FAQ
Common questions about DeepSeek-V4-Pro.
When was DeepSeek-V4-Pro released?
DeepSeek-V4-Pro's default version was released on Aug 13, 2026.
How much does DeepSeek-V4-Pro cost?
DeepSeek-V4-Pro's official API price is $0.66 per million input tokens and $1.98 per million output tokens via DeepSeek. The lowest tracked third-party offer starts at $0.1452 input and $0.4356 output via NanoGPT.
Who created DeepSeek-V4-Pro?
DeepSeek-V4-Pro was created by DeepSeek.
What is the context window for DeepSeek-V4-Pro?
The default version has a 1M token context window.
Is DeepSeek-V4-Pro open weight?
Yes. The default version is marked as open weight under MIT.