Skip to content

DeepSeek · Chat

DeepSeek V4 Pro 0813

DeepSeek 🇨🇳ChatDeepSeek familyOpen weights. One of its sources says the weights are publicly available.The maker lists this model as deprecated (30 Oct 2026).
From Baseten's listing:
A 1.6T-parameter mixture-of-experts model with 1M-token context, built for agentic workflows, complex coding, and multi-step reasoning. Achieves frontier-class performance on SWE-bench, LiveCodeBench, and tool-use benchmarks while activating only 49B parameters at inference
llminfo IQ
136.8
Market median price
$1.98 blended / 1M
Max context
1M
Listings
148

Evidence and sources

🤗 deepseek-ai/DeepSeek-V4-Pro-0813 Model architecture, as its sources state it llminfo IQ 136.8, ranked among chat models. From 147 benchmarks. Plenty of data behind it.How llminfo IQ is computedCoding IQ 117.1, 95% interval 110–124. From 27 benchmark families across 7 sources. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.How Coding IQ is computedFair value (52.5). Costs per task about what its benchmark results predict. From 41 benchmarks with measured task costs. Plenty of data behind it.How Value for money worksYap score: tokens used per coding task vs. models with the same resultsChatty (58 ±9). Generates more tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. By effort: default 58 · high 62 · max 50.Median 82,224 output tokens per task at default effort.How the Yap score worksReleased 13 Aug 2026Maker HQ: CN (according to DeepSeek's own website)
All rankings: #29 on NanoGPT recently #33 on OpenRouter this month #37 on OpenRouter today #39 on OpenRouter this weekSources: Alibaba Cloud Model Studio (DashScope) Multi-Region API, Avian.io Models data, EmpirioLabs data, LiteLLM data, Models.dev data, Morpheus Inference Marketplace data, Baseten data, llminfo researchPrimary data from Alibaba Cloud Model Studio (DashScope) Multi-Region API from Avian.io Models: name from EmpirioLabs: primary mode from LiteLLM: deprecation date from Models.dev: weights availability from Morpheus Inference Marketplace: architecture from Baseten: description from llminfo research: release date
Tags from its sources:instructtoolsreasoningopen-sourcemoelong-contextcodingagenticfunction-calling

Advantages and trade-offs

Compared with 10 chat models, picked mainly by popularity and IQ

Advantages

Open weights available

The weights are public. 3 of the 9 peers that say either way have open weights.

Trade-offs

Deprecated by its maker

The maker has already deprecated this model (30 Oct 2026).

We only make a claim when there's enough data on models of the same type. Updated 22 Nov, 06:07 UTC.

Input & output

as stated by the model's providers
Text
Images
Text

Capabilities

Supported operations and provider-stated features

VisionReasoningFunction callingStructured outputPrompt cachingWeb search

Listings

 Hugging FaceInference by Relacedeepseek-ai/DeepSeek-V4-Pro-0813Source: Hugging Face Inference Providers data1M$0.0999$8.00——
 OpenRouterInference by Relacedeepseek/deepseek-v4-pro-0813 · fp4Source: OpenRouter data docs1M$0.0999$8.0063724ms
 NanoGPTInference by RelaceRegions: UNKdeepseek/deepseek-v4-pro-0813 · fp4Source: NanoGPTNot published by the provider$0.1049$8.40135.81,044ms
 Aviandeepseek/deepseek-v4-pro-0813Source: Avian.io Models data1M$0.561cache read $0.0187$1.683——
 Meganovadeepseek-ai/DeepSeek-V4-Pro-0813Source: MegaNova data1M$0.60$1.80——
 Iteracomputedeepseek/deepseek-v4-pro-0813Source: Models.dev data1M$0.60cache read $0.044$1.79——
 Runinfradeepseek-ai/DeepSeek-V4-Pro-0813Source: Models.dev data1M$0.60cache read $0.03$1.90——
 Eden AIInference by Alibaba Cloudqwen/deepseek-v4-pro-0813Source: Models.dev data1M$0.66cache read $0.066$1.98——
 Wavespeeddeepseek/deepseek-v4-pro-0813Source: WaveSpeed AI data1M$0.66cache read $0.022$1.98——
 Ollamadeepseek-v4-pro:0813Source: Models.dev data1M$0.66cache read $0.022$1.98——

9 listings without a price are hidden.

Benchmarks

Named effort levels run from lowest to highest. Select a result to see its source details.

Agentic

Reported by DeepSeek · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
DeepSeek V4.1 launch chart · settings unspecified
official scaffold · pass@1
BFCL-V4Atria Dawn Preview official card / report Table 1 · published comparison run · model snapshots and reasoning tiers not specified unless stated · official BFCL V4 harness commit 6ea5797 · 5,217 cases · temperature 0.001 · weighted overall score · Reported by Shanghai AI Laboratory · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
BrowseCompAtria Dawn Preview official card / report Table 1 · published comparison run · model snapshots and reasoning tiers not specified unless stated · in-house search/visit/python harness · 500 steps · discard-all context strategy · Reported by Shanghai AI Laboratory · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
DeepResearch Bench IIAtria Dawn Preview official card / report Table 1 · published comparison run · model snapshots and reasoning tiers not specified unless stated · Claude Code · identical search tools · four-hour task limit · GPT-5.5 judge · three independent scoring passes · Reported by Shanghai AI Laboratory · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
GDPval · Atria paper EloAtria Dawn Preview official card / report Table 1 · published comparison run · model snapshots and reasoning tiers not specified unless stated · 220 tasks · AgentCompass / OpenHands · four-hour limit · 250 steps · GLM-5.3 rubric judge · pairwise Elo · Reported by Shanghai AI Laboratory · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
GDPval-AA v2Artificial Analysis / Z.ai转载 · Reported by Z.ai · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
JobBenchAtria Dawn Preview official card / report Table 1 · published comparison run · model snapshots and reasoning tiers not specified unless stated · official OpenCode harness · main split / 65 tasks · 7,200-second limit · two attempts · Grok-4.3 judge · mean task score · Reported by Shanghai AI Laboratory · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
SWE-Bench ProAtria Dawn Preview official card / report Table 1 · published comparison run · model snapshots and reasoning tiers not specified unless stated · OpenHands v1.23.0 · 4.5-hour limit · 250 interaction steps · network disabled and Git history removed · Reported by Shanghai AI Laboratory · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Source: OpenRouter benchmark exports
13 Aug 2026 version · 90-day results
Across 27 providers lowest 70.3% median 76.6% highest 78.6%OpenRouter's automatic routing 77.3%
Show results from 27 providers and OpenRouter's automatic routingHide results from 27 providers and OpenRouter's automatic routing
  • DeepInfra
  • Novita
  • AtlasCloud
  • Alibaba
  • Reka
  • OpenRouter's automatic routing
  • Relace
  • Baidu
  • SiliconFlow
  • Venice
  • StreamLake
  • Wafer
  • Phala
  • Parasail
  • DekaLLM
  • NextBit
  • Pearl Research
  • Together
  • CoreWeave
  • Cloudflare
  • Sail Research
  • BaseTen
  • Ionstream
  • GMICloud
  • Fireworks
  • Nebius
  • Io Net
  • DigitalOcean
Terminal-Bench3 related benchmarks · 6 results
Terminal-Bench 2.1
Reported by Z.ai · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Setting not stated
Reported by Shanghai AI Laboratory · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Atria Dawn Preview official card / report Table 1 · published comparison run · model snapshots and reasoning tiers not specified unless stated · four-run mean · 12× timeout multiplier capped at 12 hours · 500 turns · Claude Code v2.1.224 except GPT models use Terminus 2
Reported by DeepSeek · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
DeepSeek V4.1 launch chart · settings unspecified
Source: Hugging Face Official Benchmarks data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
Terminal-Bench 3.0
Reported by DeepSeek · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
DeepSeek V4.1 launch chart · settings unspecified
Terminal-Bench 4.0
Reported by DeepSeek · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
DeepSeek V4.1 launch chart · settings unspecified

Science

GPQA Diamond29 results
Reported by DeepSeek · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
DeepSeek V4.1 launch chart · settings unspecified
Pass@1 / max effort
Source: OpenRouter benchmark exports
13 Aug 2026 version · 90-day results
Across 25 providers lowest 75.4% median 90.6% highest 91.9%OpenRouter's automatic routing 90.9%
Show results from 25 providers and OpenRouter's automatic routingHide results from 25 providers and OpenRouter's automatic routing
  • NextBit
  • StreamLake
  • SiliconFlow
  • Baidu
  • AtlasCloud
  • Alibaba
  • Cloudflare
  • Wafer
  • Phala
  • DekaLLM
  • Novita
  • OpenRouter's automatic routing
  • Reka
  • GMICloud
  • Relace
  • BaseTen
  • Sail Research
  • Parasail
  • Fireworks
  • Ionstream
  • Nebius
  • CoreWeave
  • DigitalOcean
  • Together
  • Venice
  • DeepInfra

Long context

Coding

Codeforces RatingDeepSeek V4.1 launch chart · settings unspecified · Reported by DeepSeek · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
CyberGym4 results
Reported by Shanghai AI Laboratory · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Atria Dawn Preview official card / report Table 1 · published comparison run · model snapshots and reasoning tiers not specified unless stated · Claude Code v2.1.119 · high thinking effort · four-hour-ten-minute limit · 2 CPUs / 4 GB
Reported by Z.ai · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Claude Code / 无Web工具
Reported by DeepSeek · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
DeepSeek V4.1 launch chart · settings unspecified
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
DeepSWESource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Reported by DeepSeek · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
official model-card setting
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
DSBench-Hard2 results
Reported by DeepSeek · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
official model-card setting
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
NL2RepoSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
SEC-Bench ProDeepSeek V4.1 launch chart · settings unspecified · Reported by DeepSeek · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.

Reasoning

Reported by Z.ai · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Setting not stated
Reported by DeepSeek · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
DeepSeek V4.1 launch chart · settings unspecified
public subset · pass@1
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
AutomationBench v1.0.6Atria Dawn Preview official card / report Table 1 · published comparison run · model snapshots and reasoning tiers not specified unless stated · v1.0.6 with null-type fix from PR #13 · candidate harness and task count unstated · Reported by Shanghai AI Laboratory · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Humanity's Last Exam4 related benchmarks · 8 results
Humanity's Last Exam
Reported by DeepSeek · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
DeepSeek V4.1 launch chart · settings unspecified · text-only subset
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
(no tools, text-only)
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
(with tools, text-only)
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
(with tools)
Reported by DeepSeek · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
DeepSeek V4.1 launch chart · settings unspecified
with tools · pass@1
Reported by Z.ai · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.

General

Math

MathArena ApexDeepSeek V4.1 launch chart · settings unspecified · Reported by DeepSeek · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.

Factuality

Knowledge

Other

Claw-Eval v1.1Reported by Claw-Eval · via BenchBoard docs
DeepSWE v1.13 results
Reported by DeepSeek · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
DeepSeek V4.1 launch chart · settings unspecified
mini-SWE / DeepSeek设置
Reported by Z.ai · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
mini-swe-agent / Z.ai设置
MLE-Bench Lite · HumanRankAtria Dawn Preview official card / report Table 1 · published comparison run · model snapshots and reasoning tiers not specified unless stated · Claude Code · 12-hour task limit · HumanRank score · Reported by Shanghai AI Laboratory · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
NL2Repo-Bench3 results
Reported by Z.ai · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Setting not stated
Reported by DeepSeek · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
DeepSeek V4.1 launch chart · settings unspecified · launch-page table
SkillsBench8 related benchmarks · 8 results
Text-only (79 tasks)
Reported by Tencent Hy · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Hy4 official Benchmark Appendix · highest available reasoning setting · Claude Code · 79 self-contained text-only tasks · three-run average
Text-only (79 tasks) · OpenHands
Reported by Shanghai AI Laboratory · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Atria Table 1 / Appendix B · OpenHands within AgentCompass · 79 self-contained text-only tasks · three-run average · reasoning tiers and task-source commit not stated
Vals · With Skills · Accuracy
Reported by Vals AI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=deepseek/deepseek-v4-pro-0813 · provider=DeepSeek · reasoning_effort=max · max_output_tokens=384000
Vals · With Skills · Mean cost / test
Reported by Vals AI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=deepseek/deepseek-v4-pro-0813 · provider=DeepSeek · reasoning_effort=max · max_output_tokens=384000
Vals · With Skills · Mean duration / test
Reported by Vals AI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=deepseek/deepseek-v4-pro-0813 · provider=DeepSeek · reasoning_effort=max · max_output_tokens=384000
Vals · Without Skills · Accuracy
Reported by Vals AI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · Without Skills · system=deepseek/deepseek-v4-pro-0813 · provider=DeepSeek · reasoning_effort=max · max_output_tokens=384000
Show 2 more resultsHide 2 results
Vals · Without Skills · Mean cost / test
Reported by Vals AI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · Without Skills · system=deepseek/deepseek-v4-pro-0813 · provider=DeepSeek · reasoning_effort=max · max_output_tokens=384000
Vals · Without Skills · Mean duration / test
Reported by Vals AI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · Without Skills · system=deepseek/deepseek-v4-pro-0813 · provider=DeepSeek · reasoning_effort=max · max_output_tokens=384000
Workspace-Bench · English Full · Atria runAtria Dawn Preview official card / report Table 1 · published comparison run · model snapshots and reasoning tiers not specified unless stated · AgentCompass · English Full · two-hour limit · 2 CPUs / 8 GiB · Seed 2.0 Lite judge · GPT in Codex, others in Claude Code · metric subtype unstated · Reported by Shanghai AI Laboratory · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Workspace-Bench · English Lite · Atria runAtria Dawn Preview official card / report Table 1 · published comparison run · model snapshots and reasoning tiers not specified unless stated · AgentCompass · English Lite · two-hour limit · 2 CPUs / 8 GiB · Seed 2.0 Lite judge · GPT in Codex, others in Claude Code · metric subtype unstated · Reported by Shanghai AI Laboratory · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
τ³-Bench · Banking Knowledge · Atria runAtria Dawn Preview official card / report Table 1 · published comparison run · model snapshots and reasoning tiers not specified unless stated · 97 Banking Knowledge tasks · modified Tau2/Tau3 runner from commit a2c0247 · alltools BM25/dense/shell retrieval · GPT-5.2 low judge and simulator · 200-step limit · Reported by Shanghai AI Laboratory · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.

Data credits: BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links. · Data by LLM Stats · Epoch AI Benchmarking Hub, CC BY 4.0 (licence) · BullshitBench by Peter Gostev, MIT (licence) · OpenRouter public benchmark leaderboard and search-lane exports, CC BY 4.0. Fractional source scores converted to percentages; immutable dated snapshots. (licence) · Hugging Face official benchmark leaderboards; licence varies by benchmark

Also known as: @cf/deepseek-ai/deepseek-v4-pro-0813, baseten/deepseek-ai/deepseek-v4-pro-0813, databricks/databricks-deepseek-v4-pro-0813, deepinfra/deepseek-ai/deepseek-v4-pro-0813, deepseek v4 pro 0813, deepseek-ai/deepseek-v4-pro-0813, deepseek-v4-pro-0813@eu, deepseek-v4-pro:0813, deepseek/deepseek-v4-pro-0813, deepseek/deepseek-v4-pro-0813:thinking, empiriolabs/deepseek-v4-pro-0813, fireworks_ai/accounts/fireworks/models/deepseek-v4-pro-0813, fireworks_ai/deepseek-v4-pro-0813, gmi/deepseek-ai/deepseek-v4-pro-0813, nebius/deepseek-ai/deepseek-v4-pro-0813, novita/deepseek/deepseek-v4-pro-0813, parasail-deepseek-v4-pro-0813, parasail-deepseek-v4-pro-0813-fast, perplexity/deepseek-v4-pro-0813, qwen/deepseek-v4-pro-0813, streamlake/deepseek-v4-pro-0813, tencent/deepseek-v4-pro-0813, tensorx/deepseek/deepseek-v4-pro-0813, together_ai/deepseek-ai/deepseek-v4-pro-0813, wandb/deepseek-ai/deepseek-v4-pro-0813

Like this one. Tap or hover any dotted value, label or icon to see where it comes from and how it is measured. open details when tapped or hovered. How this data works