GPT-5.2 vs Mistral Medium 3.1
Updated October 21, 2026. 70 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.
Decision reading
Only GPT-5.2 has a coding score, 55.9 on the Vibe Coding Index; Mistral Medium 3.1 has none, so the two cannot be ranked on coding.
Mistral Medium 3.1 has the edge where output tokens cost 7.0x less, and where what matters is LMArena Text: software and IT or LMArena Text: hard prompts in English, where it leads.
- Shared results
- 70
- GPT-5.2 only
- 41
- Mistral Medium 3.1 only
- 4
- Like-for-like categories
- 8 of 9
Shared results can support a head-to-head reading. Results only one model has describe coverage, not superiority.
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
No call
Only GPT-5.2 has a coding score, so no call is made.
One side not measured
Intelligence
AA Intelligence Index
GPT-5.2
GPT-5.2 leads, 30.4 against 9.2.
Same board, both measured
Agentic work
AA Agentic Index
No call
Only Mistral Medium 3.1 has a AA Agentic Index result, 1.9.
Directional only
Reasoning
GPQA Diamond
No call
Only GPT-5.2 has a GPQA Diamond result, 91.7%.
Directional only
Math
AIME
GPT-5.2
GPT-5.2 leads, 96.9% against 42.3%.
Same board, both measured
Knowledge
MMLU Pro
GPT-5.2
GPT-5.2 leads, 86.2% against 75.3%.
Same board, both measured
Web design
Design Arena: websites
GPT-5.2
GPT-5.2 leads, 1195 against 1133.
Same board, both measured
Long documents
Listed context window
GPT-5.2
GPT-5.2 takes 400K tokens against 131K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Mistral Medium 3.1
GPT-5.2 $0.0088, Mistral Medium 3.1 $0.0014 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
No call
Only GPT-5.2 has the number, $1.91.
Directional only
At a glance
- Coding index
- GPT-5.2 55.9 from 3 benchmarks, Mistral Medium 3.1 not benchmarked.
- Widest category
- Agentic work, 44.2 points apart on average over 1 shared benchmark.
- Widest single gap
- AA Math Index: GPT-5.2 99.0 against Mistral Medium 3.1 38.3, 60.7 points apart on a 0 to 100 scale.
- Output price
- GPT-5.2 $14.00 against Mistral Medium 3.1 $2.00 per million output tokens, 7.0x apart.
- Context window
- GPT-5.2 takes 400K tokens, Mistral Medium 3.1 131K.
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-5.2 | Bars | Mistral Medium 3.1 | Ties |
|---|---|---|---|---|
| Coding | 5 | 1 | ||
| Intelligence | 1 | 0 | ||
| Agentic work | 1 | 0 | ||
| Reasoning | 7 | 1 | ||
| Math | 6 | 0 | ||
| Knowledge | 7 | 4 | ||
| Web design | 7 | 1 | ||
| General | 18 | 11 | ||
| All shared boards | 52 | 18 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index51
| Benchmark | GPT-5.2 | Mistral Medium 3.1 | Leader |
|---|---|---|---|
| SWE-bench Verified | 72.8% | Not measured | Only GPT-5.2 measured |
| SWE-bench Multilingual | 66.7% | Not measured | Only GPT-5.2 measured |
| LMArena WebDev | 1417 | Not measured | Only GPT-5.2 measured |
| AA Coding Index | Not measured | 20.5 | Only Mistral Medium 3.1 measured |
| LiveCodeBench (Vals AI) | 85.4% | 44.8% | GPT-5.2 |
| SWE-bench Verified (Vals AI) | 75.8% | Not measured | Only GPT-5.2 measured |
| Vibe Code Bench (Vals AI) | 53.5% | Not measured | Only GPT-5.2 measured |
| LiveCodeBench (Artificial Analysis run) | 89.4% | 40.6% | GPT-5.2 |
| SciCode | Not measured | 32.4% | Only Mistral Medium 3.1 measured |
| Terminal-Bench Hard (Artificial Analysis run) | 47.0% | 10.6% | GPT-5.2 |
| Terminal-Bench | 64.9% | Not measured | Only GPT-5.2 measured |
| GSO-Bench | 27.5% | Not measured | Only GPT-5.2 measured |
| WeirdML | 72.2% | Not measured | Only GPT-5.2 measured |
| ALE-Bench | 1293.5 | 210.2 | GPT-5.2 |
| LMArena Text: coding | 1442 | 1433 | GPT-5.2 |
| LMArena Text: software and IT | 1434 | 1436 | Mistral Medium 3.1 |
| LMArena WebDev: HTML | 1427 | Not measured | Only GPT-5.2 measured |
| Neither model has a result on the other 21 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index10
| Benchmark | GPT-5.2 | Mistral Medium 3.1 | Leader |
|---|---|---|---|
| AA Intelligence Index | 30.4 | 9.2 | GPT-5.2 |
| Epoch Capabilities Index | 153.7 | Not measured | Only GPT-5.2 measured |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index10
| Benchmark | GPT-5.2 | Mistral Medium 3.1 | Leader |
|---|---|---|---|
| AA Agentic Index | Not measured | 1.9 | Only Mistral Medium 3.1 measured |
| Design Arena: agents building web apps | 1086 | Not measured | Only GPT-5.2 measured |
| Design Arena: agents building full-stack apps | 1043 | Not measured | Only GPT-5.2 measured |
| Design Arena: agents building mobile apps | 1093 | Not measured | Only GPT-5.2 measured |
| Design Arena: agents building Android apps | 1066 | Not measured | Only GPT-5.2 measured |
| Design Arena: agents building Godot games | 1142 | Not measured | Only GPT-5.2 measured |
| τ²-Bench Telecom | 84.8% | 40.6% | GPT-5.2 |
| METR time horizons | 75.3% | Not measured | Only GPT-5.2 measured |
| Remote Labor Index | 2.5% | Not measured | Only GPT-5.2 measured |
| GDPval | 49.7% | Not measured | Only GPT-5.2 measured |
| DeepResearch Bench | 41.1% | Not measured | Only GPT-5.2 measured |
| Vending-Bench 2 | $3,591 | Not measured | Only GPT-5.2 measured |
| Neither model has a result on the other 27 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index71
| Benchmark | GPT-5.2 | Mistral Medium 3.1 | Leader |
|---|---|---|---|
| GPQA Diamond | 91.7% | Not measured | Only GPT-5.2 measured |
| MMMU Pro (Vals AI) | 86.7% | 63.0% | GPT-5.2 |
| GPQA Diamond (Artificial Analysis run) | 90.3% | 58.8% | GPT-5.2 |
| Humanity's Last Exam (Artificial Analysis run) | 37.7% | 4.7% | GPT-5.2 |
| GPQA Diamond (Epoch AI run) | 91.4% | Not measured | Only GPT-5.2 measured |
| ARC-AGI | 86.2% | Not measured | Only GPT-5.2 measured |
| ARC-AGI-2 | 52.9% | Not measured | Only GPT-5.2 measured |
| SimpleBench | 45.8% | Not measured | Only GPT-5.2 measured |
| Chess puzzles | 48.0% | Not measured | Only GPT-5.2 measured |
| Mystery game puzzles | 23.0% | Not measured | Only GPT-5.2 measured |
| EBR-bench | 23.0% | Not measured | Only GPT-5.2 measured |
| VPCT | 84.0% | Not measured | Only GPT-5.2 measured |
| LMArena Text: hard prompts | 1428 | 1426 | GPT-5.2 |
| LMArena Text: expert | 1445 | 1408 | GPT-5.2 |
| LMArena Text: hard prompts in English | 1431 | 1435 | Mistral Medium 3.1 |
| LMArena Vision: diagrams | 1265 | 1175 | GPT-5.2 |
| LMArena Vision: homework | 1276 | 1165 | GPT-5.2 |
| Neither model has a result on the other 4 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index60
| Benchmark | GPT-5.2 | Mistral Medium 3.1 | Leader |
|---|---|---|---|
| AIME | 96.9% | 42.3% | GPT-5.2 |
| MATH 500 (Vals AI) | Not measured | 87.0% | Only Mistral Medium 3.1 measured |
| MGSM (Vals AI) | 94.0% | 90.9% | GPT-5.2 |
| AA Math Index | 99.0 | 38.3 | GPT-5.2 |
| AIME 2025 (Artificial Analysis run) | 99.0% | 38.3% | GPT-5.2 |
| OTIS Mock AIME | 96.1% | Not measured | Only GPT-5.2 measured |
| FrontierMath, tiers 1 to 3 | 67.4% | Not measured | Only GPT-5.2 measured |
| FrontierMath, tier 4 | 31.7% | Not measured | Only GPT-5.2 measured |
| ProofBench | 15.0% | Not measured | Only GPT-5.2 measured |
| LMArena Text: math | 1440 | 1408 | GPT-5.2 |
| LMArena Text: mathematical work | 1442 | 1412 | GPT-5.2 |
| Neither model has a result on the other 2 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index74
| Benchmark | GPT-5.2 | Mistral Medium 3.1 | Leader |
|---|---|---|---|
| MMLU Pro | 86.2% | 75.3% | GPT-5.2 |
| LegalBench (Vals AI) | 82.8% | 62.0% | GPT-5.2 |
| CaseLaw (Vals AI) | 66.0% | Not measured | Only GPT-5.2 measured |
| CorpFin (Vals AI) | 65.9% | 50.7% | GPT-5.2 |
| TaxEval (Vals AI) | 75.8% | 70.3% | GPT-5.2 |
| MortgageTax (Vals AI) | 67.1% | 36.5% | GPT-5.2 |
| MedQA (Vals AI) | 94.1% | 78.2% | GPT-5.2 |
| MedCode (Vals AI) | 49.8% | Not measured | Only GPT-5.2 measured |
| MMLU Pro (Artificial Analysis run) | 87.4% | 68.3% | GPT-5.2 |
| SimpleQA Verified | 37.1% | Not measured | Only GPT-5.2 measured |
| LMArena Text: science | 1425 | 1444 | Mistral Medium 3.1 |
| LMArena Text: medicine and healthcare | 1425 | 1449 | Mistral Medium 3.1 |
| LMArena Text: legal and government | 1427 | 1439 | Mistral Medium 3.1 |
| LMArena Text: business and finance | 1414 | 1418 | Mistral Medium 3.1 |
| Neither model has a result on the other 3 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index71
| Benchmark | GPT-5.2 | Mistral Medium 3.1 | Leader |
|---|---|---|---|
| Design Arena: websites | 1195 | 1133 | GPT-5.2 |
| Design Arena: UI components | 1190 | 1099 | GPT-5.2 |
| Design Arena: game dev | 1195 | 1074 | GPT-5.2 |
| Design Arena: data viz | 1202 | 1144 | GPT-5.2 |
| Design Arena: 3D | 1079 | 1081 | Mistral Medium 3.1 |
| Design Arena: all code categories | 1170 | 1120 | GPT-5.2 |
| Design Arena: SVG | 1139 | 993 | GPT-5.2 |
| Design Arena: ASCII art | 1152 | 997 | GPT-5.2 |
| Neither model has a result on the other 1 board here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index1811
| Benchmark | GPT-5.2 | Mistral Medium 3.1 | Leader |
|---|---|---|---|
| SAGE (Vals AI) | 49.3% | Not measured | Only GPT-5.2 measured |
| MedScribe (Vals AI) | 84.4% | Not measured | Only GPT-5.2 measured |
| IFBench | 75.4% | 39.8% | GPT-5.2 |
| Long context reasoning (AA-LCR) | 82.7% | 42.7% | GPT-5.2 |
| DTBench | 90.9% | 67.5% | GPT-5.2 |
| LMCA | 43.9% | 19.1% | GPT-5.2 |
| CL-bench | 18.2% | Not measured | Only GPT-5.2 measured |
| Furniture assembly | 38.3% | Not measured | Only GPT-5.2 measured |
| LMArena Text | 1416 | 1424 | Mistral Medium 3.1 |
| LMArena Text: instruction following | 1409 | 1398 | GPT-5.2 |
| LMArena Text: creative writing | 1376 | 1391 | Mistral Medium 3.1 |
| LMArena Vision | 1246 | 1172 | GPT-5.2 |
| LMArena Text: English | 1422 | 1438 | Mistral Medium 3.1 |
| LMArena Text: non-English | 1405 | 1408 | Mistral Medium 3.1 |
| LMArena Text: Chinese | 1447 | 1447 | Mistral Medium 3.1 |
| LMArena Text: German | 1444 | 1432 | GPT-5.2 |
| LMArena Text: Japanese | 1420 | 1379 | GPT-5.2 |
| LMArena Text: Korean | 1374 | 1380 | Mistral Medium 3.1 |
| LMArena Text: French | 1454 | 1460 | Mistral Medium 3.1 |
| LMArena Text: Spanish | 1413 | 1433 | Mistral Medium 3.1 |
| LMArena Text: Russian | 1415 | 1411 | GPT-5.2 |
| LMArena Text: Polish | 1426 | 1430 | Mistral Medium 3.1 |
| LMArena Text: multi-turn | 1419 | 1418 | GPT-5.2 |
| LMArena Text: longer queries | 1413 | 1406 | GPT-5.2 |
| LMArena Text: writing and language | 1392 | 1395 | Mistral Medium 3.1 |
| LMArena Text: entertainment and media | 1378 | 1390 | Mistral Medium 3.1 |
| LMArena Vision: English | 1243 | 1179 | GPT-5.2 |
| LMArena Vision: Chinese | 1305 | 1190 | GPT-5.2 |
| LMArena Vision: reading text | 1257 | 1175 | GPT-5.2 |
| LMArena Vision: humor | 1227 | 1168 | GPT-5.2 |
| LMArena Vision: captioning | 1267 | 1145 | GPT-5.2 |
| LMArena Vision: recognizing things | 1232 | 1147 | GPT-5.2 |
| LMArena Vision: creative writing | 1225 | 1200 | GPT-5.2 |
| LMArena Document | 1405 | Not measured | Only GPT-5.2 measured |
| Neither model has a result on the other 12 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | GPT-5.2 | Mistral Medium 3.1 | 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 Medium 3.1 costs less
Repository review
50K input + 3K output tokens
Mistral Medium 3.1 costs less
Agent loop
200K input + 10K output tokens
Mistral Medium 3.1 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-5.2 | Mistral Medium 3.1 |
|---|---|---|
| Maker | OpenAI | Mistral AI |
| Weights | No public weights listed | No public weights listed |
| Released | December 10, 2025 | August 13, 2025 |
| Context window | 400K | 131K |
| Input $/Mtok | $1.75 | $0.40 |
| Output $/Mtok | $14.00 | $2.00 |
| Coding index | 55.9 | Not benchmarked |
| Independent measurements behind it | 3 | 0 |
| Cost per solved task | $1.91 | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is GPT-5.2 better than Mistral Medium 3.1 for coding?
Only GPT-5.2 has a coding score, 55.9 on the Vibe Coding Index; Mistral Medium 3.1 has none, so the two cannot be ranked on coding. Mistral Medium 3.1 has the edge where output tokens cost 7.0x less, and where what matters is LMArena Text: software and IT or LMArena Text: hard prompts in English, where it leads.
Which is cheaper, GPT-5.2 or Mistral Medium 3.1?
GPT-5.2 costs $1.75 per million input tokens and $14.00 per million output tokens; Mistral Medium 3.1 costs $0.40 and $2.00. Cost per solved task is known for GPT-5.2 only.
Which has the bigger context window, GPT-5.2 or Mistral Medium 3.1?
GPT-5.2, with 400K tokens against 131K.
How many benchmarks is the GPT-5.2 vs Mistral Medium 3.1 comparison based on?
70 benchmarks have a result for both models as of October 21, 2026. GPT-5.2's coding index rests on 3 independent benchmarks and Mistral Medium 3.1's on 0. Where only one model has a result, the page says so instead of naming a winner.