DeepSeek-V4.1-Flash vs Mistral Large 4
Wins 8 of 8 areas
Coding · Agents · Reasoning · Facts · Images and charts · Math · Long documents · Following instructions
Wins 0 of 8 areas
—
DeepSeek-V4.1-Flash is the stronger all-rounder.
Scores updated · 38 tests both models report · How we compare
Where each one wins
Tests won in each of the eight areas we test. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.
- CodingWriting and fixing software40DeepSeek-V4.1-Flash4 of 5 tests · 1 tie
- ReasoningHard problems that need careful thinking30DeepSeek-V4.1-Flash3 of 3 tests
- MathCompetition and research-level math20DeepSeek-V4.1-Flash2 of 3 tests · 1 tie
- AgentsCarrying out multi-step tasks on its own20DeepSeek-V4.1-Flash2 of 2 tests
- Images and chartsUnderstanding pictures, charts and video20DeepSeek-V4.1-Flash2 of 2 tests
- FactsGetting facts right instead of making them up21DeepSeek-V4.1-Flash2 of 3 tests
- Long documentsFinding answers in very long texts10DeepSeek-V4.1-Flash1 of 1 test
- Following instructionsDoing exactly what it is asked10DeepSeek-V4.1-Flash1 of 1 test
Long documents and following instructions rest on a single test each.
What it costs
Prices per million tokens, roughly 750,000 words. The bars show the cost of a million tokens read plus a million written.
DeepSeek-V4.1-Flash costs 46% less for the same work.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where DeepSeek-V4.1-Flash pulls ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+20.6points ahead
- Short factual questions, answered correctlySimpleQA Verified+20.5points ahead
- Fixes real GitHub issues, working as an agentLiveBench · Agentic Coding+20.1points ahead
Where Mistral Large 4 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+54.6points ahead
Every test, side by side
All 38 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingDeepSeek-V4.1-Flash
- LiveBench · Agentic CodingDeepSeek-V4.1-Flash by 20.177.357.2+20.1
- LMArena · WebDevDeepSeek-V4.1-Flash by 48 rating points16211573+48 rating
- LiveBench · CodingDeepSeek-V4.1-Flash by 2.88077.2+2.8
- SciCodeDeepSeek-V4.1-Flash by 1.755.954.2+1.7
- Terminal-Bench 2.1tie90.690.2tie
AgentsDeepSeek-V4.1-Flash
- Terminal-Bench 4.0DeepSeek-V4.1-Flash by 9.133.824.7+9.1
- GDPValDeepSeek-V4.1-Flash by 8.85546.2+8.8
ReasoningDeepSeek-V4.1-Flash
- CritPtDeepSeek-V4.1-Flash by 142410+14
- Humanity's Last ExamDeepSeek-V4.1-Flash by 8.343.335+8.3
- LiveBench · ReasoningDeepSeek-V4.1-Flash by 2.886.783.9+2.8
FactsDeepSeek-V4.1-Flash
- AA-Omniscience · Non-hallucinationMistral Large 4 by 54.63.558.1+54.6
- AA-Omniscience · AccuracyDeepSeek-V4.1-Flash by 20.646.425.8+20.6
- SimpleQA VerifiedDeepSeek-V4.1-Flash by 20.542.922.4+20.5
Images and chartsDeepSeek-V4.1-Flash
- MMMU-ProDeepSeek-V4.1-Flash by 4.280.676.4+4.2
- LMArena · VisionDeepSeek-V4.1-Flash by 32 rating points12781246+32 rating
MathDeepSeek-V4.1-Flash
- FrontierMath Tier 4DeepSeek-V4.1-Flash by 19.546.326.8+19.5
- FrontierMath Tiers 1-3 (v2)DeepSeek-V4.1-Flash by 12.671.258.6+12.6
- LiveBench · Mathematicstie93.393.6tie
Long documentsDeepSeek-V4.1-Flash
- AA-LCRDeepSeek-V4.1-Flash by 1.482.781.3+1.4
Following instructionsDeepSeek-V4.1-Flash
- LiveBench · Instruction FollowingDeepSeek-V4.1-Flash by 5.27064.8+5.2
Other results18 tests, not counted
Tests outside the eight areas. They are not counted above: several are summary scores built from other tests, or the same test under another name.
- livebench_languageDeepSeek-V4.1-Flash by 31.681.249.6+31.6
- DeepSWE 1.1DeepSeek-V4.1-Flash by 12.574.261.7+12.5
- Terminal-Bench 4.0 (AA Coding Agent Index)Mistral Large 4 by 103141+10
- Terminal-Bench 4.0Mistral Large 4 by 9.831.241+9.8
- Terminal-Bench-Science 0.1DeepSeek-V4.1-Flash by 8.612.94.3+8.6
- HLE-DiamondDeepSeek-V4.1-Flash by 825.317.3+8
- CyberGymDeepSeek-V4.1-Flash by 6.188.182+6.1
- StrongREJECT (cyber subset): refusal rateMistral Large 4 by 5.68792.6+5.6
- livebench_data_analysisDeepSeek-V4.1-Flash by 2.879.376.5+2.8
- CyberGym-E2E-AAMistral Large 4 by 2.776.379+2.7
- JailbreakBench (cyber subset): safe-response rateMistral Large 4 by 2.794.697.3+2.7
- CritPt v2 (Hard Subset)Mistral Large 4 by 22426+2
- CritPt v2 HardMistral Large 4 by 22527+2
- AA IntelligenceDeepSeek-V4.1-Flash by 1.139.538.4+1.1
- GDP (Surge AI)tie17.217.6tie
- AA-Omnisciencetie-5.3-5.3tie
- AgentHarm (cyber subset): 100 minus harm scoretie100100tie
- BeyondAIMEtie90.690.6tie
Questions people ask
Which is better, DeepSeek-V4.1-Flash or Mistral Large 4?
DeepSeek-V4.1-Flash wins all eight areas we test: coding, agents, reasoning, facts, images and charts, math, long documents and following instructions. Mistral Large 4 wins none.
Which is better for coding?
DeepSeek-V4.1-Flash. It wins 4 of the 5 coding tests both models report; Mistral Large 4 wins none, and 1 is a tie.
Which is cheaper?
DeepSeek-V4.1-Flash costs $0.30 per million input tokens and $1.20 per million output tokens; Mistral Large 4 costs $0.70 and $2.10. That makes DeepSeek-V4.1-Flash about 46% cheaper for the same work.
How do you compare the two?
We use the 38 benchmark tests both models have published scores on. The verdict counts the 20 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 18 are listed but not counted, because several are summary scores or repeat a test. Each score is the one shown on the model's own page.