Gemini 3.8 Flash vs Mistral Large 4
Wins 4 of 8 areas
Reasoning · Facts · Images and charts · Following instructions
Wins 3 of 8 areas
Coding · Agents · Math
Gemini 3.8 Flash wins more areas, narrowly.Mistral Large 4 is cheaper and better at coding.
Scores updated · 30 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.
- ReasoningHard problems that need careful thinking30Gemini 3.8 Flash3 of 3 tests
- Images and chartsUnderstanding pictures, charts and video20Gemini 3.8 Flash2 of 2 tests
- FactsGetting facts right instead of making them up21Gemini 3.8 Flash2 of 3 tests
- Following instructionsDoing exactly what it is asked10Gemini 3.8 Flash1 of 1 test
- CodingWriting and fixing software23Mistral Large 43 of 5 tests
- MathCompetition and research-level math12Mistral Large 42 of 3 tests
- AgentsCarrying out multi-step tasks on its own01Mistral Large 41 of 2 tests · 1 tie
- Long documentsFinding answers in very long texts00Even0 each · 1 tie
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.
Mistral Large 4 costs 38% less for the same work.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where Gemini 3.8 Flash pulls ahead
- Short factual questions, answered correctlySimpleQA Verified+47.3points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+28.8points ahead
- Follows detailed instructions exactlyLiveBench · Instruction Following+16.6points ahead
Where Mistral Large 4 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+13.3points ahead
- Complex command-line tasks across many fieldsTerminal-Bench 4.0+5points ahead
- Extremely hard research-level math problemsFrontierMath Tier 4+4.8points ahead
Every test, side by side
All 30 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingMistral Large 4
- LiveBench · CodingMistral Large 4 by 4.772.577.2+4.7
- LiveBench · Agentic CodingMistral Large 4 by 354.257.2+3
- Terminal-Bench 2.1Mistral Large 4 by 2.687.690.2+2.6
- SciCodeGemini 3.8 Flash by 2.456.654.2+2.4
- LMArena · WebDevGemini 3.8 Flash by 14 rating points15841570+14 rating
AgentsMistral Large 4
- Terminal-Bench 4.0Mistral Large 4 by 519.724.7+5
- GDPValtie46.846.2tie
ReasoningGemini 3.8 Flash
- Humanity's Last ExamGemini 3.8 Flash by 12.847.835+12.8
- CritPtGemini 3.8 Flash by 8.318.310+8.3
- LiveBench · ReasoningGemini 3.8 Flash by 5.489.383.9+5.4
FactsGemini 3.8 Flash
- SimpleQA VerifiedGemini 3.8 Flash by 47.369.722.4+47.3
- AA-Omniscience · AccuracyGemini 3.8 Flash by 28.854.625.8+28.8
- AA-Omniscience · Non-hallucinationMistral Large 4 by 13.344.858.1+13.3
Images and chartsGemini 3.8 Flash
- MMMU-ProGemini 3.8 Flash by 9.285.676.4+9.2
- LMArena · VisionGemini 3.8 Flash by 73 rating points13171244+73 rating
MathMistral Large 4
- FrontierMath Tiers 1-3 (v2)Gemini 3.8 Flash by 9.868.458.6+9.8
- FrontierMath Tier 4Mistral Large 4 by 4.821.926.8+4.8
- LiveBench · MathematicsMistral Large 4 by 291.693.6+2
Following instructionsGemini 3.8 Flash
- LiveBench · Instruction FollowingGemini 3.8 Flash by 16.681.464.8+16.6
Other results10 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_languageGemini 3.8 Flash by 38.287.849.6+38.2
- AA-OmniscienceGemini 3.8 Flash by 34.929.6-5.3+34.9
- livebench_data_analysisMistral Large 4 by 22.55476.5+22.5
- Terminal-Bench 4.0Mistral Large 4 by 21.919.141+21.9
- CritPt v2 HardGemini 3.8 Flash by 19.546.527+19.5
- HLE-DiamondGemini 3.8 Flash by 15.833.117.3+15.8
- DeepSWE 1.1Gemini 3.8 Flash by 1273.761.7+12
- Terminal-Bench-Science 0.1Gemini 3.8 Flash by 5.7104.3+5.7
- GDP (Surge AI)Gemini 3.8 Flash by 3.42117.6+3.4
- AA IntelligenceGemini 3.8 Flash by 2.540.938.4+2.5
Questions people ask
Which is better, Gemini 3.8 Flash or Mistral Large 4?
Gemini 3.8 Flash wins four of the eight areas we test: reasoning, facts, images and charts and following instructions. Mistral Large 4 wins coding, agents and math, and costs 38% less. They are level on long documents.
Which is better for coding?
Mistral Large 4. It wins 3 of the 5 coding tests both models report; Gemini 3.8 Flash wins 2.
Which is cheaper?
Gemini 3.8 Flash costs $0.75 per million input tokens and $3.75 per million output tokens; Mistral Large 4 costs $0.70 and $2.10. That makes Mistral Large 4 about 38% cheaper for the same work.
How do you compare the two?
We use the 30 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 10 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.