Model comparison
GPT-5 Nano vs Mistral Large 3
Mistral Large 3 is the stronger model overall, scoring 38.6 to 33.4 on the Noometry Index. GPT-5 Nano costs 2.7× less per token, which makes it the better buy when Mistral Large 3's lead doesn't matter for your workload.
Last verified . 19 shared benchmarks.
Mistral AISummary
- They share 19 benchmarks with published results for both. GPT-5 Nano scores higher in 3 categories and Mistral Large 3 in 6 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Mistral Large 3 leads 60.2 to 39.4.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 62.2% for GPT-5 Nano and 50.9% for Mistral Large 3.
- GPT-5 Nano is cheaper at $0.05 / $0.40 per million input/output tokens, against $0.25 / $0.75 for Mistral Large 3.
- GPT-5 Nano accepts more context: 400K tokens versus 262K.
- Mistral Large 3 has downloadable open weights; the other is API-only.
Side by side
| GPT-5 Nano | Mistral Large 3 | |
|---|---|---|
| Provider | OpenAI | Mistral AI |
| Noometry Index | 33.4 | 38.6 |
| Released | 2025-08-07 | 2025-12-02 |
| Weights | Proprietary | Open |
| Context window | 400K | 262K |
| Max output | 128K | 8K |
| Input $ / M tokens | $0.05 | $0.25 |
| Output $ / M tokens | $0.40 | $0.75 |
| Results tracked | 48 | 24 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
GPT-5 Nano: 33.3 (#248), Mistral Large 3: 34.1 (#230)
| Benchmark | GPT-5 Nano | Mistral Large 3 |
|---|---|---|
| LMArena Coding | 1351 | 1447 |
| SWE-bench Verified (bash only) | 34.8% | — |
| LMArena WebDev | — | 1230 |
| WeirdML | 38.1% | — |
| ALE-Bench | 718.67 | — |
Agentic & Tool Use Not comparable
GPT-5 Nano: 25.6 (#99), Mistral Large 3: —
| Benchmark | GPT-5 Nano | Mistral Large 3 |
|---|---|---|
| Terminal-Bench | 21.8% | — |
| Berkeley Function Calling Leaderboard | 51.5% | — |
Reasoning GPT-5 Nano leads
GPT-5 Nano: 16.4 (#291), Mistral Large 3: 14.6 (#314)
| Benchmark | GPT-5 Nano | Mistral Large 3 |
|---|---|---|
| Kagi LLM Benchmark | 62.2% | 50.9% |
| LMArena Hard Prompts | 1329 | 1429 |
| ARC-AGI-2 | 2.6% | — |
| NYT Connections (extended) | — | 7.5% |
| ARC-AGI-1 | 20.7% | — |
| Chess Puzzles | 27% | — |
| Thematic Generalization | — | 23% |
| Mystery Game Puzzles | 9% | — |
| DTBench | 62.7% | — |
| LMCA | 7.9% | — |
| Epoch Capabilities Index | 139.39 | — |
| ForecastBench | 59.1 | — |
Math Mistral Large 3 leads
GPT-5 Nano: 29.1 (#221), Mistral Large 3: 36.0 (#131)
| Benchmark | GPT-5 Nano | Mistral Large 3 |
|---|---|---|
| LMArena Math | 1318 | 1415 |
| FrontierMath (Tiers 1-3) | 20% | — |
| FrontierMath Tier 4 | 2.4% | — |
| OTIS Mock AIME 2024-2025 | 81.1% | — |
| ProofBench | 12% | — |
| Omni-MATH | 54.6% | — |
| MATH Level 5 | 95.2% | — |
| FrontierMath (Feb 2025 set) | 8.3% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge Too close to call
GPT-5 Nano: 35.8 (#166), Mistral Large 3: 35.5 (#173)
| Benchmark | GPT-5 Nano | Mistral Large 3 |
|---|---|---|
| Vectara Hallucination Rate | 10.5% | 14.5% |
| LMArena Expert | 1321 | 1421 |
| GPQA Diamond | 69.4% | — |
| SimpleQA Verified | 11.7% | — |
| MMLU-Pro | 77.8% | — |
| GPQA (HELM) | 67.9% | — |
Multimodal Mistral Large 3 leads
GPT-5 Nano: 30.7 (#102), Mistral Large 3: 37.6 (#65)
| Benchmark | GPT-5 Nano | Mistral Large 3 |
|---|---|---|
| LMArena Vision | 1159 | 1222 |
| VPCT | 37.2% | — |
Multilingual Mistral Large 3 leads
GPT-5 Nano: 45.5 (#167), Mistral Large 3: 52.6 (#79)
| Benchmark | GPT-5 Nano | Mistral Large 3 |
|---|---|---|
| LMArena Non-English | 1313 | 1412 |
| LMArena Chinese | 1355 | 1448 |
| LMArena German | 1329 | 1438 |
| LMArena Korean | 1269 | 1384 |
| LMArena Russian | 1296 | 1412 |
| LMArena Spanish | 1360 | 1437 |
| LMArena French | — | 1456 |
| LMArena Japanese | — | 1399 |
Instruction Following GPT-5 Nano leads
GPT-5 Nano: 75.2 (#72), Mistral Large 3: 74.1 (#104)
| Benchmark | GPT-5 Nano | Mistral Large 3 |
|---|---|---|
| LMArena Instruction Following | 1306 | 1402 |
| IFEval | 93.2% | — |
Long Context Mistral Large 3 leads
GPT-5 Nano: 31.3 (#275), Mistral Large 3: 43.1 (#102)
| Benchmark | GPT-5 Nano | Mistral Large 3 |
|---|---|---|
| LMArena Longer Query | 1313 | 1413 |
| Fiction.LiveBench | 44.4% | — |
Writing & Preference Mistral Large 3 leads
GPT-5 Nano: 39.4 (#244), Mistral Large 3: 60.2 (#100)
| Benchmark | GPT-5 Nano | Mistral Large 3 |
|---|---|---|
| LMArena Text | 1320 | 1427 |
| LMArena Creative Writing | 1250 | 1385 |
| EQ-Bench Creative Writing | 705 | 1412 |
| LMArena Multi-Turn | 1311 | 1429 |
| WildBench | 80.6% | — |
Frequently asked questions
Is GPT-5 Nano better than Mistral Large 3?
Mistral Large 3 is the stronger model overall, scoring 38.6 to 33.4 on the Noometry Index. GPT-5 Nano costs 2.7× less per token, which makes it the better buy when Mistral Large 3's lead doesn't matter for your workload.
Which is cheaper, GPT-5 Nano or Mistral Large 3?
GPT-5 Nano is cheaper. It lists at $0.05 per million input tokens and $0.40 per million output tokens; Mistral Large 3 lists at $0.25 and $0.75.
Is GPT-5 Nano or Mistral Large 3 better for coding?
They score almost the same on coding (33.3 vs 34.1); test both on your own repository before choosing.
Which has the bigger context window?
GPT-5 Nano does, with 400K tokens against 262K.
How many benchmarks do GPT-5 Nano and Mistral Large 3 share?
19 benchmarks have published results for both models. GPT-5 Nano has 48 scored results on Noometry and Mistral Large 3 has 24.