Skip to content
AI Atlas
ModelAnnouncedOpen weightsIdentity probable

DeepSeek V3.1

DeepSeekfamily · DeepSeekchat.deepseek.com

DeepSeek-V3.1 is a large hybrid reasoning model (671B parameters, 37B active) that supports both thinking and non-thinking modes via prompt templates. It extends the DeepSeek-V3 base with a two-phase long-context...

Open in Graph
data quality72

Updated 5 h ago · first seen 11 Sept 2026

model_01M294WWCNME74MRVVP3J0Y4C3

Overview

Identity

Canonical model
Yesidentity confidence: mediumOne row per real model release. Artifacts (checkpoints, quantisations, conversions) and folded evaluation variants point here.
Official checkpoints
official_checkpoints = hf_repo identifiers carried by the model itself; artifacts are separate entities pointing here through canonical_id.
Artifacts
None recordedSeparate entities (checkpoint · quantization · conversion · packaging) pointing to this model through canonical_id.
Provider deployments
2
API aliases
deepseek-v3-1deepseek-v3-1-reasoningdeepseek/deepseek-chat-v3.1Identifiers under which providers and evaluators refer to this model.
Folded evaluation variants
2Effort / thinking variants (…-high, …-non-reasoning) are result configurations of this model, not separate models. Their old URLs redirect here.

Openness

Open weightsweights downloadable; 7 dimensions unknown.

Weights downloadable under a permissive or Creative Commons licence allowing commercial use; code or data may be missing.

  • Weights

    Yes

  • Inference code

  • Training code

  • Training data

  • Dataset

  • Commercial use

  • Redistribution

  • Derivatives

dimensions marked null are unknown, not false

Key facts

Release date

Source:DeepSeek — site & API docsT2observed 15 h agomediumLLM-extracted

Status

Source:DeepSeek — site & API docsT2observed 15 h agomediumLLM-extracted

Version

Source:DeepSeek — site & API docsT2observed 15 h agomediumLLM-extracted

Knowledge cutoff

Source:OpenRouter public model & pricing listingT2observed 15 h agomedium

Official page

Source:DeepSeek — site & API docsT2observed 15 h agomediumLLM-extracted

Repository

Source:DeepSeek — site & API docsT2observed 15 h agomediumLLM-extracted

Openrouter id

Source:OpenRouter public model & pricing listingT2observed 15 h agomedium

Architecture

Architecture

Source:DeepSeek — site & API docsT2observed 15 h agomediumLLM-extracted

Parameters

Source:DeepSeek — site & API docsT2observed 15 h agomediumLLM-extracted

Tokenizer

Source:OpenRouter public model & pricing listingT2observed 15 h agomedium

Hugging Face repo

Source:OpenRouter public model & pricing listingT2observed 15 h agomedium

Capabilities

Modalities

Modalities
text
Input
text
Output
text

Capabilities

  • Tool calling

    Yes

    OpenRouter public model & pricing listing · T2

  • Structured output

    Yes

    OpenRouter public model & pricing listing · T2

  • Reasoning

    Yes

    OpenRouter public model & pricing listing · T2

  • Vision

    Unavailable

  • Audio

    Unavailable

  • Fine-tuning available

    Unavailable

Context window

Source:OpenRouter public model & pricing listingT2observed 7 h agomedium

Max output

Source:OpenRouter public model & pricing listingT2observed 15 h agomedium

Knowledge cutoff

Source:OpenRouter public model & pricing listingT2observed 15 h agomedium

Tokenizer

Source:OpenRouter public model & pricing listingT2observed 15 h agomedium

Comparable same task and conditions · Partially comparable same task, conditions differ (effort, temperature, judge) · Not comparable different variant or metric

Benchmark results grouped by comparability group
Benchmark · groupBest scoreTrustConfigurationResultsvs leaderEvaluatedSource
Terminal-Benchagentic · accuracy · variant=hard · evaluator=Artificial AnalysisIndependentevaluatorArtificial Analysisvarianthardreasoningonconditions differ across rows → partially comparable4−40.9 ptvs gpt-5.6-solobs. 12 Sept 2026artificialanalysis.aiT2
τ²-benchagentic · pass^1 · variant=Telecom · evaluator=Artificial AnalysisIndependentevaluatorArtificial AnalysisvariantTelecomreasoningonconditions differ across rows → partially comparable2non-primary groupobs. 12 Sept 2026artificialanalysis.aiT2
τ²-benchagentic · pass^1 · evaluator=Artificial AnalysisIndependentevaluatorArtificial Analysisreasoningonconditions differ across rows → partially comparable2−61.7 ptvs Z.ai GLM 5.2obs. 11 Sept 2026artificialanalysis.aiT2
Artificial Analysis Intelligence Indexcomposite · indexIndependentreasoningoffversion4.3conditions differ across rows → partially comparable4−39.7vs Claude Fable 5.1obs. 12 Sept 2026artificialanalysis.aiT2
IFBenchinstruction-following · accuracy · evaluator=Artificial AnalysisIndependentevaluatorArtificial Analysisreasoningonconditions differ across rows → partially comparable4−41.8 ptvs Grok 4.3obs. 12 Sept 2026artificialanalysis.aiT2
Humanity's Last Examknowledge · accuracy · evaluator=Artificial AnalysisIndependentevaluatorArtificial Analysisreasoningonconditions differ across rows → partially comparable4−44.9 ptvs Claude Fable 5.1obs. 12 Sept 2026artificialanalysis.aiT2
GPQA Diamondreasoning · accuracy · variant=Diamond · evaluator=Artificial AnalysisIndependentevaluatorArtificial AnalysisvariantDiamondreasoningonconditions differ across rows → partially comparable2non-primary groupobs. 12 Sept 2026artificialanalysis.aiT2
GPQA Diamondreasoning · accuracy · variant=GPQA Diamond · evaluator=Artificial AnalysisIndependentevaluatorArtificial AnalysisvariantGPQA Diamondreasoningonconditions differ across rows → partially comparable2−18.4 ptvs gpt-6-astraobs. 11 Sept 2026artificialanalysis.aiT2

Current rows only, grouped by benchmark → canonical metric → comparability group (task configuration). Effort variants folded into this model appear as rows of the same group. 24 current rows in total. “vs leader” compares with the current leader of the benchmark's primary group only; other groups are not directly comparable. Comparability rules →

Provider deployments, cheapest output first
ProviderContextInput / 1MCached inOutput / 1MStatusObservedSource
DeepSeek APIcheapest outputdeepseek/deepseek-chat-v3.1163.8Kout 32.8K$0.13active6 h agosince 11 Sept 2026openrouter.aiT2
OpenRouterdeepseek/deepseek-chat-v3.1163.8Kout 32.8K$0.13active2 h agosince 12 Sept 2026openrouter.aiT2

USD per 1M tokens as published by each provider; native units (per-request fees, flex/priority tiers) are kept verbatim. Rows are append-only — every price change is kept in the history below. Cost of a workload →

Price history

Step lines per provider; amber markers are recorded changes. Click a marker or a row for the evidence behind that price.

Output price · USD / 1M tokens 2 providers

Output price history of DeepSeek V3.1$0$0.20$0.40$0.60$0.80$1$1.2Sept 26Sept 26Sept 26Sept 26Sept 26DeepSeek API: first observed → $0.95 · 11 Sept 2026OpenRouter: first observed → $0.95 · 12 Sept 2026
  • DeepSeek API
  • OpenRouter
  • OpenRouterfirst observed $0.9512 Sept 2026
  • DeepSeek APIfirst observed $0.9511 Sept 2026

Input price · USD / 1M tokens 2 providers

Input price history of DeepSeek V3.1$0$0.10$0.20$0.30Sept 26Sept 26Sept 26Sept 26Sept 26DeepSeek API: first observed → $0.25 · 11 Sept 2026OpenRouter: first observed → $0.25 · 12 Sept 2026
  • DeepSeek API
  • OpenRouter
  • OpenRouterfirst observed $0.2512 Sept 2026
  • DeepSeek APIfirst observed $0.2511 Sept 2026

Hardware fit37

Estimated

3 of 37 device × quantization combinations fit.

Run locally: your machine →
Estimated hardware fit
HardwareQuantizationDevice memoryEst. memoryFits
Apple M3 Ultra4bit483.5 GB est.Yes
Mac Studio (Apple M5 Ultra)4bit483.5 GB est.Yes
NVIDIA DGX B2004bit1,440 GB483.5 GB est.Yes
Apple M2 Ultra4bit483.5 GB est.No
Apple M1 Ultra4bit483.5 GB est.No
Apple M3 Max4bit483.5 GB est.No
Apple M4 Max4bit483.5 GB est.No
Mac Studio (Apple M5 Max)4bit483.5 GB est.No
MacBook Pro (Apple M5 Max)4bit483.5 GB est.No
Apple M2 Max4bit483.5 GB est.No
Apple M1 Max4bit483.5 GB est.No
Apple M4 Pro4bit483.5 GB est.No
Mac mini (Apple M5 Pro)4bit483.5 GB est.No
MacBook Pro (Apple M5 Pro)4bit483.5 GB est.No
Apple M3 Pro4bit483.5 GB est.No
Apple M1 Pro4bit483.5 GB est.No
Apple M2 Pro4bit483.5 GB est.No
Apple M44bit483.5 GB est.No
iMac (Apple M4)4bit483.5 GB est.No
Mac mini (Apple M6)4bit483.5 GB est.No
MacBook Air (Apple M5)4bit483.5 GB est.No
MacBook Pro (Apple M5)4bit483.5 GB est.No
Apple M24bit483.5 GB est.No
Apple M34bit483.5 GB est.No
Apple M14bit483.5 GB est.No
NVIDIA GeForce RTX 30904bit24 GB483.5 GB est.No
NVIDIA GeForce RTX 40904bit24 GB483.5 GB est.No
NVIDIA GeForce RTX 50904bit32 GB483.5 GB est.No
NVIDIA A100 80GB4bit80 GB483.5 GB est.No
NVIDIA H100 SXM4bit80 GB483.5 GB est.No
NVIDIA H100 NVL4bit94 GB483.5 GB est.No
NVIDIA DGX Spark4bit128 GB483.5 GB est.No
NVIDIA H2004bit141 GB483.5 GB est.No
NVIDIA H200 NVL4bit141 GB483.5 GB est.No
NVIDIA B2004bit180 GB483.5 GB est.No
AMD Instinct MI300X4bit192 GB483.5 GB est.No
AMD Instinct MI325X4bit256 GB483.5 GB est.No
Assumptions (7)
  • Estimated, not measured: weights = parameters × bytes/param × 1.15 runtime overhead (or the observed artifact file size when one is recorded).
  • bytes/param: 4bit = 0.5, 8bit = 1.0, fp16 = 2.0 (uniform quantization, no per-layer exceptions).
  • KV cache: 2 × layers × kv_heads × head_dim × 2 bytes × context × batch when the architecture is known; otherwise 0.5 GB per 8 192 tokens (× batch), independent of architecture (GQA/MLA models need less).
  • A model 'fits' when the estimate is at most the device memory minus 2 GB reserved for the OS and framework.
  • Mixture-of-experts models are estimated on total parameters (all experts must be resident); active parameters are ignored.
  • Device memory uses the largest configuration when several are listed (e.g. Apple silicon tiers).
  • Multi-GPU: device memories are summed; interconnect bandwidth, tensor-parallel replication and pipeline bubbles are not modelled.

Versions & Artifacts0

Version history

Context window3 changes

11 Sept 202611 Sept 202611 Sept 202612 Sept 2026

Status1 change

11 Sept 202611 Sept 2026

Knowledge cutofffirst observation only

11 Sept 2026current

Max outputfirst observation only

11 Sept 2026current

Opennessfirst observation only

11 Sept 2026current

Parametersfirst observation only

11 Sept 2026current

Each hop is a claim: click a value for its source, tier and observation time. Nothing is overwritten — a new observation closes the previous claim.

Artifacts 0

No artifact (checkpoint, quantisation, conversion or packaging) points to this model yet.

Change history

Temporal, append-only claims: a new observation closes the previous claim instead of overwriting it. Rewind the record with the as-of picker.
3 claims · 1 propertiesShow all properties

Statusstatus3

Claim history for Status
ValueValid from → toStatusSourceConfidenceExtractor
deprecatedsupersededArtificial AnalysisT2mediumdeterministic
deprecatedcurrentcurrentArtificial AnalysisT2mediumdeterministic
announcedcurrentcurrentDeepSeek — site & API docsT2mediumllm

Claims are temporal and append-only: a new observation closes the previous claim (valid_to) instead of overwriting it. Conflicting claims from different sources are kept side by side and flagged — never averaged. Methodology →

Provenance

Attributed facts

28

Source tiers

T228

Freshest observation

5 h ago

Conflicts

None

Source documents 3

Source documents
SourceDocumentTypeTierLast observedSnapshots
DeepSeek — site & API docsapi-docs.deepseek.com/news/news250821 newsT1· Official15 h ago1
OpenRouter public model & pricing listingopenrouter.ai/api/v1/models catalogueT2· Quality secondary2 h ago11
Artificial Analysisartificialanalysis.ai/leaderboards/models leaderboardT2· Quality secondary5 h ago2

Tier 1 = official/primary, 2 = quality secondary, 3 = community, 4 = unverified. Every snapshot is archived; see all sources and the methodology.

Data quality (72/100) measures how well AI Atlas knows this entity — completeness, primary-source ratio, freshness, conflicts — never how good the model is. Methodology →