Translate
Independent model intelligence

Find the right AI model,
not just the loudest one.

Compare current model capability, context, modalities, public preference and API economics in one continuously refreshed decision surface.

LiveBench 2.0 LMArena models.dev OpenRouter AIHOT cross-check
Back to leaderboard
DeepSeek · Active

deepseek-v4.1-flash-max

deepseek-flash family · launched 2026-09-10

MultimodalReasoningToolsStructuredOpen weights
AITopic score76.0#13 overall

DeepSeek V4.1 Flash model for reasoning and agentic coding

Evidence coverage100%8 evaluations · 3 institutions
Context / maximum output1M / 384ktokens
Cache / input / output$0.003 · $0.15 · $0.6USD per million tokens
Value / freshness89.6 / 88relative value · release freshness
Product profile

What this model supports

Inputtext, image
Outputtext
Knowledge cutoff2025-05
WeightsOpen weights
Capability profile

Strengths by domain

Coding75.42 public results
Research & reasoning86.72 public results
Professional work70.71 public results
Knowledge & instruction73.52 public results
Behind the score

Every result has a source

Raw results and standardized readings are shown together. Missing tests add neither points nor penalties.

Coding2 results

LiveBench 2.0 · CodingLiveBench90.20 scoreReading 90.2

Imported from LiveBench 2.0. The published 0–100 score is used directly; it is not z-score normalized.

Configuration: LiveBench 2.0, AIHOT aggregateOpen evidence
AIHOT · CodingAIHOT62.00 scoreReading 62.0

Imported from AIHOT aggregate leaderboard. The published 0–100 score is used directly; it is not z-score normalized.

Configuration: AIHOT aggregateOpen evidence

Research & reasoning2 results

LiveBench 2.0 · ResearchLiveBench86.70 scoreReading 86.7

Imported from LiveBench 2.0. The published 0–100 score is used directly; it is not z-score normalized.

Configuration: LiveBench 2.0, AIHOT aggregateOpen evidence
AIHOT · ResearchAIHOT59.70 scoreReading 59.7

Imported from AIHOT aggregate leaderboard. The published 0–100 score is used directly; it is not z-score normalized.

Configuration: LiveBench 2.0, AIHOT aggregateOpen evidence

Professional work1 results

LiveBench 2.0 · ProfessionalLiveBench70.65 scoreReading 70.7

Imported from LiveBench 2.0. The published 0–100 score is used directly; it is not z-score normalized. Professional = long-agentic, browser, data-analysis and design average.

Configuration: LiveBench 2.0, AIHOT aggregateOpen evidence

Knowledge & instruction2 results

LiveBench 2.0 · KnowledgeLiveBench86.80 scoreReading 86.8

Imported from LiveBench 2.0. The published 0–100 score is used directly; it is not z-score normalized. Knowledge uses the published LiveBench research suite.

Configuration: LiveBench 2.0, AIHOT aggregateOpen evidence
AIHOT · KnowledgeAIHOT62.10 scoreReading 62.1

Imported from AIHOT aggregate leaderboard. The published 0–100 score is used directly; it is not z-score normalized.

Configuration: AIHOT aggregateOpen evidence

Human preference & indices4 results

LiveBenchLiveBench78.04 scoreReading 78.0

Imported from LiveBench 2.0. The published 0–100 score is used directly; it is not z-score normalized.

Configuration: LiveBench 2.0, AIHOT aggregateOpen evidence
Arena TextLMArena1,463.5 EloReading 62.0

Direct LMArena text leaderboard result.

Open evidence
AIHOT aggregate leaderboardAIHOT61.80 scoreReading 61.8

Imported from AIHOT aggregate leaderboard. The published 0–100 score is used directly; it is not z-score normalized.

Configuration: AIHOT aggregateOpen evidence
AA IndexArtificial Analysis66.07 indexReading 61.6

Imported from Artificial Analysis. The published 0–100 score is used directly; it is not z-score normalized.

Configuration: AIHOT aggregateOpen evidence
Same builder

More from DeepSeek