GPT-6 Astra vs Mistral Large 4
Wins 7 of 8 areas
Coding · Agents · Reasoning · Facts · Images and charts · Math · Following instructions
Wins 0 of 8 areas
—
GPT-6 Astra is the stronger all-rounder.Mistral Large 4 is cheaper.
Scores updated · 31 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.
- AgentsCarrying out multi-step tasks on its own30GPT-6 Astra3 of 3 tests
- ReasoningHard problems that need careful thinking30GPT-6 Astra3 of 3 tests
- MathCompetition and research-level math30GPT-6 Astra3 of 3 tests
- CodingWriting and fixing software31GPT-6 Astra3 of 5 tests · 1 tie
- Images and chartsUnderstanding pictures, charts and video20GPT-6 Astra2 of 2 tests
- FactsGetting facts right instead of making them up21GPT-6 Astra2 of 3 tests
- Following instructionsDoing exactly what it is asked10GPT-6 Astra1 of 1 test
- 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 95% 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 GPT-6 Astra pulls ahead
- Extremely hard research-level math problemsFrontierMath Tier 4+70.8points ahead
- Short factual questions, answered correctlySimpleQA Verified+53.2points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+36.8points ahead
Where Mistral Large 4 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+9.4points ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+1.8points ahead
Every test, side by side
All 31 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGPT-6 Astra
- LMArena · WebDevGPT-6 Astra by 193 rating points17631570+193 rating
- LiveBench · CodingGPT-6 Astra by 3.280.477.2+3.2
- SciCodeGPT-6 Astra by 2.356.554.2+2.3
- Terminal-Bench 2.1Mistral Large 4 by 1.888.490.2+1.8
- LiveBench · Agentic Codingtie57.357.2tie
AgentsGPT-6 Astra
- Terminal-Bench 4.0GPT-6 Astra by 34.459.124.7+34.4
- GDPValGPT-6 Astra by 7.453.646.2+7.4
- BrowseCompGPT-6 Astra by 791.584.5+7
ReasoningGPT-6 Astra
- CritPtGPT-6 Astra by 21.731.710+21.7
- Humanity's Last ExamGPT-6 Astra by 19.754.735+19.7
- LiveBench · ReasoningGPT-6 Astra by 8.892.783.9+8.8
FactsGPT-6 Astra
- SimpleQA VerifiedGPT-6 Astra by 53.275.622.4+53.2
- AA-Omniscience · AccuracyGPT-6 Astra by 36.862.625.8+36.8
- AA-Omniscience · Non-hallucinationMistral Large 4 by 9.448.758.1+9.4
Images and chartsGPT-6 Astra
- MMMU-ProGPT-6 Astra by 10.586.976.4+10.5
- LMArena · VisionGPT-6 Astra by 35 rating points12791244+35 rating
MathGPT-6 Astra
- FrontierMath Tier 4GPT-6 Astra by 70.897.626.8+70.8
- FrontierMath Tiers 1-3 (v2)GPT-6 Astra by 35.193.758.6+35.1
- LiveBench · MathematicsGPT-6 Astra by 3.296.893.6+3.2
Following instructionsGPT-6 Astra
- LiveBench · Instruction FollowingGPT-6 Astra by 10.875.664.8+10.8
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.
- Terminal-Bench-Science 0.1GPT-6 Astra by 5963.34.3+59
- AA-OmniscienceGPT-6 Astra by 48.743.4-5.3+48.7
- HLE-DiamondGPT-6 Astra by 46.563.817.3+46.5
- CritPt v2 HardGPT-6 Astra by 43.570.527+43.5
- livebench_languageGPT-6 Astra by 39.889.449.6+39.8
- Terminal-Bench 4.0GPT-6 Astra by 16.757.741+16.7
- AA IntelligenceGPT-6 Astra by 14.352.738.4+14.3
- GDP (Surge AI)GPT-6 Astra by 13.43117.6+13.4
- DeepSWE 1.1GPT-6 Astra by 12.474.161.7+12.4
- livebench_data_analysisGPT-6 Astra by 6.58376.5+6.5
Questions people ask
Which is better, GPT-6 Astra or Mistral Large 4?
GPT-6 Astra wins seven of the eight areas we test: coding, agents, reasoning, facts, images and charts, math and following instructions. Mistral Large 4 wins none, but costs 95% less. They are level on long documents.
Which is better for coding?
GPT-6 Astra. It wins 3 of the 5 coding tests both models report; Mistral Large 4 wins 1, and 1 is a tie.
Which is cheaper?
GPT-6 Astra costs $10.00 per million input tokens and $50.00 per million output tokens; Mistral Large 4 costs $0.70 and $2.10. That makes Mistral Large 4 about 95% cheaper for the same work.
How do you compare the two?
We use the 31 benchmark tests both models have published scores on. The verdict counts the 21 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.