Skip to content
AI Atlas
ModelActiveOpen weightsIdentity probable

DeepSeek-V4.1-Flash

DeepSeekfamily · DeepSeekapi-docs.deepseek.com/quick_start/pricing

DeepSeek 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...

Open in Graph
data quality70

Updated 2 h ago · first seen 11 Sept 2026

model_01M293WR6HM862PR1R3X4523QG

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
3
API aliases
deepseek-flashdeepseek-v4-1-flashdeepseek/deepseek-v4.1-flashfireworks/deepseek-v4p1-flashIdentifiers under which providers and evaluators refer to this model.
Folded evaluation variants
0Effort / thinking variants (…-high, …-non-reasoning) are result configurations of this model, not separate models. Their old URLs redirect here.

Openness

Open weights— weights 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:OpenRouter public model & pricing listingT2observed 12 h agomedium

Official page
api-docs.deepseek.com/quick_start/pricing

Source:DeepSeek — site & API docsT1observed 12 h agohigh

API model id

Source:DeepSeek — site & API docsT1observed 12 h agohigh

Openrouter id

Source:OpenRouter public model & pricing listingT2observed 12 h agomedium

Architecture

Tokenizer

Source:OpenRouter public model & pricing listingT2observed 12 h agomedium

Hugging Face repo

Source:OpenRouter public model & pricing listingT2observed 12 h agomedium

Capabilities

Modalities

Modalities
imagetext
Input
textimage
Output
text

Capabilities

  • Tool calling

    Yes

    DeepSeek — site & API docs · T1

  • Structured output

    Yes

    DeepSeek — site & API docs · T1

  • Reasoning

    Yes

    DeepSeek — site & API docs · T1

  • Vision

    Yes

    DeepSeek — site & API docs · T1

  • Audio

    Unavailable

  • Fine-tuning available

    Unavailable

Context window

Source:DeepSeek — site & API docsT1observed 12 h agohigh

Max output

Source:DeepSeek — site & API docsT1observed 12 h agohigh

Tokenizer

Source:OpenRouter public model & pricing listingT2observed 12 h agomedium

Benchmarks18

Compare with another model →

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=v4.0 · evaluator=Artificial AnalysisIndependentevaluatorArtificial Analysisvariantv4.0reasoning_effortmaxconditions differ across rows → partially comparable2non-primary groupobs. 12 Sept 2026artificialanalysis.aiT2
LiveBench Agentic Codingcoding · average score · variant=Agentic CodingOfficial boardvariantAgentic Codingrelease2026-06-251current leader25 Jun 2026livebench.aiT2
LiveBench Codingcoding · average score · variant=CodingOfficial boardvariantCodingrelease2026-06-251−6.34 ptvs Claude Fable 5.125 Jun 2026livebench.aiT2
SciCodecoding · accuracy · evaluator=Artificial AnalysisIndependentevaluatorArtificial Analysisreasoning_effortmaxconditions differ across rows → partially comparable2−11.2 ptvs Claude Fable 5.1obs. 12 Sept 2026artificialanalysis.aiT2
Artificial Analysis Intelligence Indexcomposite · indexIndependentreasoning_effortmaxversion4.3conditions differ across rows → partially comparable2−13.8vs Claude Fable 5.1obs. 12 Sept 2026artificialanalysis.aiT2
LiveBenchgeneral · global_averageOfficial boardrelease2026-06-251−2.30 ptvs Claude Fable 5.125 Jun 2026livebench.aiT2
LiveBench Data Analysisgeneral · average score · variant=Data AnalysisOfficial boardvariantData Analysisrelease2026-06-251−3.72 ptvs gpt-6-astra25 Jun 2026livebench.aiT2
LiveBench Languagegeneral · average score · variant=LanguageOfficial boardvariantLanguagerelease2026-06-251−9.49 ptvs Claude Fable 525 Jun 2026livebench.aiT2
LiveBench Instruction Followinginstruction-following · average score · variant=IFOfficial boardvariantIFrelease2026-06-251−11.4 ptvs Gemini 3.8 Flash25 Jun 2026livebench.aiT2
Humanity's Last Examknowledge · accuracy · evaluator=Artificial AnalysisIndependentevaluatorArtificial Analysisreasoning_effortmaxconditions differ across rows → partially comparable2−19.9 ptvs Claude Fable 5.1obs. 12 Sept 2026artificialanalysis.aiT2
LiveBench Mathematicsmath · average score · variant=MathematicsOfficial boardvariantMathematicsrelease2026-06-251−3.72 ptvs Claude Fable 5.125 Jun 2026livebench.aiT2
MMMU-Promultimodal · accuracy · evaluator=Artificial AnalysisIndependentevaluatorArtificial Analysisreasoning_effortmaxconditions differ across rows → partially comparable2−9.89 ptvs gpt-6-astraobs. 12 Sept 2026artificialanalysis.aiT2
LiveBench Reasoningreasoning · average score · variant=ReasoningOfficial boardvariantReasoningrelease2026-06-251−5.96 ptvs gpt-6-astra25 Jun 2026livebench.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. 18 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 →

Providers & Pricing4

All offers in the price terminal →
Provider deployments, cheapest output first
ProviderContextInput / 1MCached inOutput / 1MNative unitsStatusObservedSource
DeepSeek APIcheapest outputdeepseek/deepseek-v4.1-flash1.05Mout 384K$0.003—active3 h agosince 11 Sept 2026openrouter.aiT2
OpenRouterdeepseek/deepseek-v4.1-flash1.05Mout 384K$0.003—active2 h agosince 12 Sept 2026openrouter.aiT2
Fireworks AIfireworks/deepseek-v4p1-flash—$0.007—active5 h agosince 11 Sept 2026app.fireworks.aiT1
DeepSeek APIdeepseek-flash—$0.006off_peak_input_per_mtok=0.15off_peak_output_per_mtok=0.6off_peak_cached_input_per_mtok=0.003active12 h agosince 11 Sept 2026api-docs.deepseek.comT1

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 3 providers

Output price history of DeepSeek-V4.1-Flash$0$0.50$1$1.5Sept 26Sept 26Sept 26Sept 26DeepSeek API: first observed → $1.2 · 11 Sept 2026Fireworks AI: first observed → $0.66 · 11 Sept 2026DeepSeek API: $1.2 → $0.60 · 11 Sept 2026OpenRouter: first observed → $0.60 · 12 Sept 2026
  • DeepSeek API
  • Fireworks AI
  • OpenRouter
  • OpenRouterfirst observed $0.6012 Sept 2026
  • DeepSeek API$1.2 → $0.6011 Sept 2026
  • Fireworks AIfirst observed $0.6611 Sept 2026
  • DeepSeek APIfirst observed $1.211 Sept 2026

Input price · USD / 1M tokens 3 providers

Input price history of DeepSeek-V4.1-Flash$0$0.10$0.20$0.30$0.40Sept 26Sept 26Sept 26Sept 26DeepSeek API: first observed → $0.30 · 11 Sept 2026Fireworks AI: first observed → $0.22 · 11 Sept 2026DeepSeek API: $0.30 → $0.15 · 11 Sept 2026OpenRouter: first observed → $0.15 · 12 Sept 2026
  • DeepSeek API
  • Fireworks AI
  • OpenRouter
  • OpenRouterfirst observed $0.1512 Sept 2026
  • DeepSeek API$0.30 → $0.1511 Sept 2026
  • Fireworks AIfirst observed $0.2211 Sept 2026
  • DeepSeek APIfirst observed $0.3011 Sept 2026

Versions & Artifacts0

Version history

Context windowfirst observation only

11 Sept 2026current

Max outputfirst observation only

11 Sept 2026current

Opennessfirst 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.

Timeline6

Full timeline →

Change history36

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

Viewing AI Atlas as of 11 Sept 2026 — attributes exactly as the atlas knew them on that day; later corrections are not shown.

Back to today →

Attributes as of 11 Sept 2026 21 claims in force

Family
DeepSeek
Release date
10 Sept 2026
Openness
open-weights
Context window
1M tokens
Max output
384K tokens
Modalities
image, text
Input modalities
text, image
Output modalities
text
Tokenizer
DeepSeek
API model id
deepseek-flash
Hugging Face repo
deepseek-ai/DeepSeek-V4.1-Flash
Livebench hf link
https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash
Aa median output tokens per second
213.2
Openrouter id
deepseek/deepseek-v4.1-flash
Reasoning
Yes
Structured output
Yes
Supported parameters
frequency_penalty, include_reasoning, logit_bias, logprobs, 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
Tool calling
Yes
Vision
Yes
36 claims · 25 properties11 conflicting

Familyfamily1

Claim history for Family
ValueValid from → toStatusSourceConfidenceExtractor
DeepSeek → currentcurrentDeepSeek — site & API docsT1highdeterministic

Release daterelease_date1

Claim history for Release date
ValueValid from → toStatusSourceConfidenceExtractor
10 Sept 2026 → currentcurrentOpenRouter public model & pricing listingT2mediumdeterministic

Opennessopenness1

Claim history for Openness
ValueValid from → toStatusSourceConfidenceExtractor
open-weights → currentcurrentArtificial AnalysisT2mediumdeterministic

Context windowcontext_length8conflicting claims

Claim history for Context window
ValueValid from → toStatusSourceConfidenceExtractor
1.05M tokens → currentconflictingOpenRouter public model & pricing listingT2mediumdeterministic
1.05M tokens → currentconflictingOpenRouter public model & pricing listingT2mediumdeterministic
1.05M tokens → currentconflictingOpenRouter public model & pricing listingT2mediumdeterministic
1.05M tokens → currentconflictingOpenRouter public model & pricing listingT2mediumdeterministic
1.05M tokens → currentconflictingOpenRouter public model & pricing listingT2mediumdeterministic
1.05M tokens → currentconflictingOpenRouter public model & pricing listingT2mediumdeterministic
1.05M tokens → currentconflictingOpenRouter public model & pricing listingT2mediumdeterministic
1M tokens → currentcurrentDeepSeek — site & API docsT1conflicteddeterministic

Max outputmax_output_tokens1

Claim history for Max output
ValueValid from → toStatusSourceConfidenceExtractor
384K tokens → currentcurrentDeepSeek — site & API docsT1highdeterministic

Modalitiesmodalities1

Claim history for Modalities
ValueValid from → toStatusSourceConfidenceExtractor
image, text → currentcurrentOpenRouter public model & pricing listingT2mediumdeterministic

Input modalitiesmodalities_input5conflicting claims

Claim history for Input modalities
ValueValid from → toStatusSourceConfidenceExtractor
image, text → currentconflictingOpenRouter public model & pricing listingT2mediumdeterministic
image, text → currentconflictingOpenRouter public model & pricing listingT2mediumdeterministic
image, text → currentconflictingOpenRouter public model & pricing listingT2mediumdeterministic
image, text → currentconflictingOpenRouter public model & pricing listingT2mediumdeterministic
text, image → currentcurrentDeepSeek — site & API docsT1conflicteddeterministic

Output modalitiesmodalities_output1

Claim history for Output modalities
ValueValid from → toStatusSourceConfidenceExtractor
text → currentcurrentOpenRouter public model & pricing listingT2mediumdeterministic

Tokenizertokenizer1

Claim history for Tokenizer
ValueValid from → toStatusSourceConfidenceExtractor
DeepSeek → currentcurrentOpenRouter public model & pricing listingT2mediumdeterministic

API model idapi_model_id1

Claim history for API model id
ValueValid from → toStatusSourceConfidenceExtractor
deepseek-flash → currentcurrentDeepSeek — site & API docsT1highdeterministic

Hugging Face repohf_repo1

Claim history for Hugging Face repo
ValueValid from → toStatusSourceConfidenceExtractor
deepseek-ai/DeepSeek-V4.1-Flash → currentcurrentOpenRouter public model & pricing listingT2mediumdeterministic

Official pageofficial_url1

Claim history for Official page
ValueValid from → toStatusSourceConfidenceExtractor
https://api-docs.deepseek.com/quick_start/pricing/ → currentcurrentDeepSeek — site & API docsT1highdeterministic

Aa context windowaa_context_window1

Claim history for Aa context window
ValueValid from → toStatusSourceConfidenceExtractor
1,000,000 tokens → currentcurrentArtificial AnalysisT2mediumdeterministic

Aa opennessaa_openness1

Claim history for Aa openness
ValueValid from → toStatusSourceConfidenceExtractor
open-weights → currentcurrentArtificial AnalysisT2mediumdeterministic

Descriptiondescription1

Claim history for Description
ValueValid from → toStatusSourceConfidenceExtractor
DeepSeek 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... → currentcurrentOpenRouter public model & pricing listingT2mediumdeterministic

Aa median output tokens per secondmetric.aa_median_output_tokens_per_second1

Claim history for Aa median output tokens per second
ValueValid from → toStatusSourceConfidenceExtractor
213.2 → currentcurrentArtificial AnalysisT2mediumdeterministic

Openrouter idopenrouter_id1

Claim history for Openrouter id
ValueValid from → toStatusSourceConfidenceExtractor
deepseek/deepseek-v4.1-flash → currentcurrentOpenRouter public model & pricing listingT2mediumdeterministic

Openrouter listed atopenrouter_listed_at1

Claim history for Openrouter listed at
ValueValid from → toStatusSourceConfidenceExtractor
10 Sept 2026 → currentcurrentOpenRouter public model & pricing listingT2mediumdeterministic

Reasoningreasoning1

Claim history for Reasoning
ValueValid from → toStatusSourceConfidenceExtractor
Yes → currentcurrentDeepSeek — site & API docsT1highdeterministic

Structured outputstructured_output1

Claim history for Structured output
ValueValid from → toStatusSourceConfidenceExtractor
Yes → currentcurrentDeepSeek — site & API docsT1highdeterministic

Supported parameterssupported_parameters1

Claim history for Supported parameters
ValueValid from → toStatusSourceConfidenceExtractor
frequency_penalty, include_reasoning, logit_bias, logprobs, 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 → currentcurrentOpenRouter public model & pricing listingT2mediumdeterministic

Tool callingtool_calling1

Claim history for Tool calling
ValueValid from → toStatusSourceConfidenceExtractor
Yes → currentcurrentDeepSeek — site & API docsT1highdeterministic

Visionvision1

Claim history for Vision
ValueValid from → toStatusSourceConfidenceExtractor
Yes → currentcurrentDeepSeek — site & API docsT1highdeterministic

Weights availableweights_available1

Claim history for Weights available
ValueValid from → toStatusSourceConfidenceExtractor
Yes → currentcurrentAI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2highderived

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

25

Source tiers

T1T210 / 15

Freshest observation

2 h ago

Conflicts

11 flagged

Source documents 5

Source documents
SourceDocumentTypeTierLast observedSnapshots
Fireworks AI — pricingdocs.fireworks.ai/serverless/pricing.md pricingT1· Official2 h ago2
DeepSeek — site & API docsapi-docs.deepseek.com/quick_start/pricing pricingT1· Official4 h ago1
OpenRouter public model & pricing listingopenrouter.ai/api/v1/models catalogueT2· Quality secondary2 h ago8
Artificial Analysisartificialanalysis.ai/leaderboards/models leaderboardT2· Quality secondary2 h ago2
LiveBenchlivebench.ai/table_2026_06_25.csv leaderboardT2· Quality secondary10 h ago1

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

Data quality (70/100) measures how well AI Atlas knows this entity — completeness, primary-source ratio, freshness, conflicts — never how good the model is. Methodology →