GPT-6.2 Astra vs Mistral Large 4
GPT-6.2 Astra leads Mistral Large 4 on coding, 87.0 to 61.1 on the Vibe Coding Index, and Mistral Large 4 costs 19.0x less per output token ($2.10 vs $40.00 per million). Updated November 22, 2026; 59 benchmarks have a result for both.
GPT-6.2 Astra or Mistral Large 4: which to pick
Pick GPT-6.2 Astra for coding: it scores 87.0 on the Vibe Coding Index against 61.1 for Mistral Large 4, though on 4 benchmarks against 5.
Pick Mistral Large 4 for output tokens 19.0x cheaper ($2.10 vs $40.00 per million) and work measured by LMArena Text: coding or LMArena Text: software and IT, where it leads.
- Shared results
- 59
- GPT-6.2 Astra only
- 17
- Mistral Large 4 only
- 32
- Like-for-like categories
- 8 of 9
Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.
Which one for your work
Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.
Coding
Vibe Coding Index, six public coding benchmarks
GPT-6.2 Astra
GPT-6.2 Astra scores 87.0 against 61.1.
4 and 5 benchmarks behind the scores
Intelligence
AA Intelligence Index
GPT-6.2 Astra
GPT-6.2 Astra leads, 62.1 against 26.1.
Same board, both measured
Web design
Design Arena: websites
GPT-6.2 Astra
GPT-6.2 Astra leads, 1333 against 1253.
Same board, both measured
General
Vals Index
GPT-6.2 Astra
GPT-6.2 Astra leads, 67.7% against 56.5%.
Same board, both measured
Long documents
Listed context window
GPT-6.2 Astra
GPT-6.2 Astra takes 1.05M tokens against 1.05M.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Mistral Large 4
GPT-6.2 Astra $0.028, Mistral Large 4 $0.0018 per turn.
List prices
At a glance
- Coding index
- GPT-6.2 Astra 87.0 from 4 benchmarks, Mistral Large 4 61.1 from 5 benchmarks.
- Widest category
- Math, 45.3 points apart on average over 4 shared benchmarks.
- Widest single gap
- Mystery game puzzles: GPT-6.2 Astra 95.0% against Mistral Large 4 17.0%, 78.0 points apart on a 0 to 100 scale.
- Output price
- GPT-6.2 Astra $40.00 against Mistral Large 4 $2.10 per million output tokens, 19.0x apart.
- Context window
- GPT-6.2 Astra takes 1.05M tokens, Mistral Large 4 1.05M.
Benchmark by benchmark
First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.
| Category | GPT-6.2 Astra | Bars | Mistral Large 4 | Ties |
|---|---|---|---|---|
| Coding | 10 | 2 | ||
| Intelligence | 2 | 0 | ||
| Agentic work | 6 | 2 | ||
| Reasoning | 7 | 2 | ||
| Math | 4 | 0 | ||
| Knowledge | 2 | 3 | ||
| Web design | 6 | 0 | ||
| General | 12 | 1 | ||
| All shared boards | 49 | 10 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index102
| Benchmark | GPT-6.2 Astra | Mistral Large 4 | Leader |
|---|---|---|---|
| LiveBench coding | 91.9% | 79.4% | GPT-6.2 Astra |
| LMArena WebDev | 1766 | 1573 | GPT-6.2 Astra |
| Terminal-Bench 4 (Vals AI) | 63.1% | 27.8% | GPT-6.2 Astra |
| Vibe Code Bench (Vals AI) | 89.9% | 81.6% | GPT-6.2 Astra |
| Code migration (Vals AI) | Not measured | 30.6% | Only Mistral Large 4 measured |
| IOI (Vals AI) | Not measured | 45.3% | Only Mistral Large 4 measured |
| SWE-bench Multimodal (Vals AI) | Not measured | 33.1% | Only Mistral Large 4 measured |
| Vibe Code Bench 1-100 (Vals AI) | Not measured | 14.3% | Only Mistral Large 4 measured |
| Terminal-Bench 4.0 (Artificial Analysis run) | 64.1% | 24.7% | GPT-6.2 Astra |
| FrontierCode 2.0 | 49.0% | 15.0% | GPT-6.2 Astra |
| FrontierCode | Not measured | 39.4% | Only Mistral Large 4 measured |
| MirrorCode | 67.8% | 40.0% | GPT-6.2 Astra |
| ALE-Bench | Not measured | 1213.2 | Only Mistral Large 4 measured |
| LMArena Text: coding | 1450 | 1459 | Mistral Large 4 |
| LMArena Text: software and IT | 1443 | 1459 | Mistral Large 4 |
| LMArena WebDev: React | 1770 | 1582 | GPT-6.2 Astra |
| LMArena WebDev: HTML | 1854 | 1560 | GPT-6.2 Astra |
| GameBench 2 | 97.2 | 57.9 | GPT-6.2 Astra |
| Terminal-Bench 4.0 | Not measured | 27.9% | Only Mistral Large 4 measured |
| Neither model has a result on the other 33 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | GPT-6.2 Astra | Mistral Large 4 | Leader |
|---|---|---|---|
| AA Intelligence Index | 62.1 | 26.1 | GPT-6.2 Astra |
| Epoch Capabilities Index | 172.8 | 154.6 | GPT-6.2 Astra |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index62
| Benchmark | GPT-6.2 Astra | Mistral Large 4 | Leader |
|---|---|---|---|
| Design Arena: agents building web apps | 1334 | Not measured | Only GPT-6.2 Astra measured |
| Design Arena: agents building full-stack apps | 1351 | Not measured | Only GPT-6.2 Astra measured |
| Design Arena: agents building mobile apps | 1305 | Not measured | Only GPT-6.2 Astra measured |
| Design Arena: agents building Android apps | 1257 | Not measured | Only GPT-6.2 Astra measured |
| Design Arena: spreadsheets | 1278 | Not measured | Only GPT-6.2 Astra measured |
| LiveBench research | 91.5% | 80.3% | GPT-6.2 Astra |
| LiveBench browser | 93.0% | 17.4% | GPT-6.2 Astra |
| LiveBench long agentic | 70.8% | 24.7% | GPT-6.2 Astra |
| Finance Agent (Vals AI) | Not measured | 54.7% | Only Mistral Large 4 measured |
| Tax Agent Bench (Vals AI) | 21.8% | 30.7% | Mistral Large 4 |
| Legal Research Bench (Vals AI) | 42.8% | 35.6% | GPT-6.2 Astra |
| Harvey's Legal Agent Benchmark (Vals AI) | 16.3% | 17.5% | Mistral Large 4 |
| Excel Modeling Benchmark (Vals AI) | Not measured | 56.2% | Only Mistral Large 4 measured |
| BioMysteryBench (Vals AI) | 80.7% | 67.0% | GPT-6.2 Astra |
| Terminal-Bench-Science (Vals AI) | 71.4% | Not measured | Only GPT-6.2 Astra measured |
| Vals Science Index | 68.4% | Not measured | Only GPT-6.2 Astra measured |
| Robo-bench (Vals AI) | 32.7% | Not measured | Only GPT-6.2 Astra measured |
| APEX-Agents | 72.2% | 51.6% | GPT-6.2 Astra |
| SpireClimb | 41.3% | Not measured | Only GPT-6.2 Astra measured |
| LMArena Search | 1191 | Not measured | Only GPT-6.2 Astra measured |
| Neither model has a result on the other 24 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index72
| Benchmark | GPT-6.2 Astra | Mistral Large 4 | Leader |
|---|---|---|---|
| LiveBench reasoning | 91.6% | 85.2% | GPT-6.2 Astra |
| MysteryMechanism (Vals AI) | 53.1% | 12.2% | GPT-6.2 Astra |
| HLE-Diamond (Artificial Analysis run) | 72.2% | 17.3% | GPT-6.2 Astra |
| CritPt v2, hard subset (Artificial Analysis run) | 71.5% | 27.0% | GPT-6.2 Astra |
| GPQA Diamond (Epoch AI run) | 96.8% | 88.1% | GPT-6.2 Astra |
| ARC-AGI | 98.5% | Not measured | Only GPT-6.2 Astra measured |
| ARC-AGI-2 | 95.8% | Not measured | Only GPT-6.2 Astra measured |
| SimpleBench | 77.3% | Not measured | Only GPT-6.2 Astra measured |
| Chess puzzles | 78.0% | 33.0% | GPT-6.2 Astra |
| Mystery game puzzles | 95.0% | 17.0% | GPT-6.2 Astra |
| EBR-bench | 92.4% | Not measured | Only GPT-6.2 Astra measured |
| LMArena Text: hard prompts | 1425 | 1436 | Mistral Large 4 |
| LMArena Text: expert | Not measured | 1450 | Only Mistral Large 4 measured |
| LMArena Text: hard prompts in English | 1426 | 1442 | Mistral Large 4 |
| LMArena Vision: diagrams | Not measured | 1273 | Only Mistral Large 4 measured |
| LMArena Vision: homework | Not measured | 1266 | Only Mistral Large 4 measured |
| Humanity's Last Exam (Scale run) | 66.9% | Not measured | Only GPT-6.2 Astra measured |
| GoBench | 2493 | Not measured | Only GPT-6.2 Astra measured |
| Neither model has a result on the other 8 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index40
| Benchmark | GPT-6.2 Astra | Mistral Large 4 | Leader |
|---|---|---|---|
| OTIS Mock AIME | 100.0% | 98.3% | GPT-6.2 Astra |
| FrontierMath, tiers 1 to 3 | 92.6% | 58.6% | GPT-6.2 Astra |
| FrontierMath, tier 4 | 100.0% | 26.8% | GPT-6.2 Astra |
| ProofBench | Not measured | 10.0% | Only Mistral Large 4 measured |
| LMArena Text: math | Not measured | 1445 | Only Mistral Large 4 measured |
| LMArena Text: mathematical work | Not measured | 1456 | Only Mistral Large 4 measured |
| MathArena | 92.9% | 20.7% | GPT-6.2 Astra |
| Neither model has a result on the other 7 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index23
| Benchmark | GPT-6.2 Astra | Mistral Large 4 | Leader |
|---|---|---|---|
| MedCode (Vals AI) | Not measured | 40.7% | Only Mistral Large 4 measured |
| AA-Omniscience | 50.0 | 5.1 | GPT-6.2 Astra |
| SimpleQA Verified | 78.5% | 20.0% | GPT-6.2 Astra |
| LMArena Text: science | 1415 | 1434 | Mistral Large 4 |
| LMArena Text: medicine and healthcare | Not measured | 1403 | Only Mistral Large 4 measured |
| LMArena Text: legal and government | 1415 | 1427 | Mistral Large 4 |
| LMArena Text: business and finance | 1413 | 1430 | Mistral Large 4 |
| Neither model has a result on the other 14 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index60
| Benchmark | GPT-6.2 Astra | Mistral Large 4 | Leader |
|---|---|---|---|
| Design Arena: websites | 1333 | 1253 | GPT-6.2 Astra |
| Design Arena: UI components | 1341 | 1272 | GPT-6.2 Astra |
| Design Arena: game dev | 1501 | 1190 | GPT-6.2 Astra |
| Design Arena: data viz | Not measured | 1271 | Only Mistral Large 4 measured |
| Design Arena: 3D | 1430 | 1225 | GPT-6.2 Astra |
| Design Arena: all code categories | 1389 | 1247 | GPT-6.2 Astra |
| Design Arena: SVG | Not measured | 1216 | Only Mistral Large 4 measured |
| Design Arena: ASCII art | Not measured | 1196 | Only Mistral Large 4 measured |
| LiveBench design | 82.8% | 76.3% | GPT-6.2 Astra |
GeneralLike-for-likePublic boards, shown as published, kept out of the index121
| Benchmark | GPT-6.2 Astra | Mistral Large 4 | Leader |
|---|---|---|---|
| Vals Index | 67.7% | 56.5% | GPT-6.2 Astra |
| LiveBench data analysis | 84.1% | 76.6% | GPT-6.2 Astra |
| MedScribe (Vals AI) | Not measured | 80.4% | Only Mistral Large 4 measured |
| DTBench | Not measured | 88.2% | Only Mistral Large 4 measured |
| LMCA | Not measured | 35.8% | Only Mistral Large 4 measured |
| Surface Evolver Bench | Not measured | 28.1% | Only Mistral Large 4 measured |
| Furniture assembly | 89.2% | 20.8% | GPT-6.2 Astra |
| LMArena Text | 1414 | 1413 | GPT-6.2 Astra |
| LMArena Text: instruction following | 1423 | 1414 | GPT-6.2 Astra |
| LMArena Text: creative writing | 1417 | 1341 | GPT-6.2 Astra |
| LMArena Vision | Not measured | 1246 | Only Mistral Large 4 measured |
| LMArena Text: English | 1417 | 1423 | Mistral Large 4 |
| LMArena Text: non-English | 1404 | 1400 | GPT-6.2 Astra |
| LMArena Text: Chinese | Not measured | 1466 | Only Mistral Large 4 measured |
| LMArena Text: French | Not measured | 1443 | Only Mistral Large 4 measured |
| LMArena Text: Spanish | Not measured | 1456 | Only Mistral Large 4 measured |
| LMArena Text: Russian | 1437 | 1415 | GPT-6.2 Astra |
| LMArena Text: multi-turn | 1425 | 1423 | GPT-6.2 Astra |
| LMArena Text: longer queries | 1431 | 1409 | GPT-6.2 Astra |
| LMArena Text: writing and language | 1440 | 1367 | GPT-6.2 Astra |
| LMArena Text: entertainment and media | 1413 | 1358 | GPT-6.2 Astra |
| LMArena Vision: English | Not measured | 1258 | Only Mistral Large 4 measured |
| LMArena Vision: Chinese | Not measured | 1331 | Only Mistral Large 4 measured |
| LMArena Vision: reading text | Not measured | 1261 | Only Mistral Large 4 measured |
| LMArena Vision: creative writing | Not measured | 1213 | Only Mistral Large 4 measured |
| MultiChallenge | 73.2% | Not measured | Only GPT-6.2 Astra measured |
| Neither model has a result on the other 23 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | GPT-6.2 Astra | Mistral Large 4 | Leader |
|---|---|---|---|
| Neither model has a result on 5 boards here | |||
What each workload costs
Three fixed token mixes at each model's listed price, so per-token rates become a bill.
Chat turn
1K input + 500 output tokens
Mistral Large 4 costs less
Repository review
50K input + 3K output tokens
Mistral Large 4 costs less
Agent loop
200K input + 10K output tokens
Mistral Large 4 costs less
Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.
Specification differences
The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.
| Specification | OpenAI GPT-6.2 Astra | Mistral Large 4 |
|---|---|---|
| Maker | OpenAI | Mistral AI |
| Weights | No public weights listed | Open weights |
| Released | November 17, 2026 | October 6, 2026 |
| Context window | 1.05M | 1.05M |
| Input $/Mtok | $8.00 | $0.70 |
| Output $/Mtok | $40.00 | $2.10 |
| Coding index | 87.0 | 61.1 |
| Independent measurements behind it | 4 | 5 |
| Cost per solved task | Not computable | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is GPT-6.2 Astra better than Mistral Large 4 for coding?
Pick GPT-6.2 Astra for coding: it scores 87.0 on the Vibe Coding Index against 61.1 for Mistral Large 4, though on 4 benchmarks against 5. Pick Mistral Large 4 for output tokens 19.0x cheaper ($2.10 vs $40.00 per million) and work measured by LMArena Text: coding or LMArena Text: software and IT, where it leads.
Which is cheaper, GPT-6.2 Astra or Mistral Large 4?
GPT-6.2 Astra costs $8.00 per million input tokens and $40.00 per million output tokens; Mistral Large 4 costs $0.70 and $2.10. Cost per solved task is known for neither of them.
Which has the bigger context window, GPT-6.2 Astra or Mistral Large 4?
GPT-6.2 Astra, with 1.05M tokens against 1.05M.
How many benchmarks is the GPT-6.2 Astra vs Mistral Large 4 comparison based on?
59 benchmarks have a result for both models as of November 22, 2026. GPT-6.2 Astra's coding index rests on 4 independent benchmarks and Mistral Large 4's on 5. Where only one model has a result, the page says so instead of naming a winner.