Model comparison
GPT-4.5 vs Mistral Large
GPT-4.5 is the stronger model overall, scoring 37.2 to 31.9 on the Noometry Index.
Last verified . 32 shared benchmarks.
Mistral AISummary
- They share 32 benchmarks with published results for both. GPT-4.5 scores higher in 7 categories and Mistral Large in 2 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where GPT-4.5 leads 56.9 to 40.7.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 37.8% for GPT-4.5 and 8.5% for Mistral Large.
- Mistral Large has downloadable open weights; the other is API-only.
Side by side
| GPT-4.5 | Mistral Large | |
|---|---|---|
| Provider | OpenAI | Mistral AI |
| Noometry Index | 37.2 | 31.9 |
| Released | 2025-02-27 | 2024-02-26 |
| Weights | Proprietary | Open |
| Context window | — | 131K |
| Max output | — | 16K |
| Input $ / M tokens | — | $2 |
| Output $ / M tokens | — | $6 |
| Results tracked | 42 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-4.5 leads
GPT-4.5: 42.2 (#109), Mistral Large: 34.3 (#240)
| Benchmark | GPT-4.5 | Mistral Large |
|---|---|---|
| LiveBench Coding | 75.2% | 47.1% |
| LMArena Coding | 1396 | 1277 |
| Aider Polyglot | 44.9% | — |
| SciCode | — | 36.2% |
| WeirdML | 39.4% | — |
| BigCodeBench Instruct | — | 30% |
| BigCodeBench Complete | — | 38.3% |
| ALE-Bench | — | 264.7 |
| HumanEval+ | — | 62.2% |
| MBPP+ | — | 59.5% |
Agentic & Tool Use Too close to call
GPT-4.5: 27.9 (#97), Mistral Large: 28.6 (#89)
| Benchmark | GPT-4.5 | Mistral Large |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 38.4% |
| Cybench | 17.5% | — |
Reasoning Mistral Large leads
GPT-4.5: 13.9 (#330), Mistral Large: 15.8 (#310)
| Benchmark | GPT-4.5 | Mistral Large |
|---|---|---|
| SimpleBench | 34.5% | 22.5% |
| LiveBench Reasoning | 71.1% | 43.5% |
| LMArena Hard Prompts | 1403 | 1257 |
| LiveBench Data Analysis | 64.3% | 50.1% |
| Epoch Capabilities Index | 136.74 | 128.52 |
| ForecastBench | 61.7 | 57.1 |
| LiveBench | 69% | 48.4% |
| ARC-AGI-2 | 0.8% | — |
| ARC-AGI-1 | 10.3% | — |
| CritPt | — | 0% |
| EnigmaEval | 3.2% | — |
| DTBench | — | 65.1% |
| LMCA | — | 16.7% |
Math GPT-4.5 leads
GPT-4.5: 32.6 (#211), Mistral Large: 18.2 (#291)
| Benchmark | GPT-4.5 | Mistral Large |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 37.8% | 8.5% |
| LiveBench Math | 69.3% | 42.5% |
| LMArena Math | 1412 | 1262 |
| MATH Level 5 | 78.6% | 50.3% |
| Omni-MATH | — | 28.1% |
| FrontierMath (Feb 2025 set) | — | 0.3% |
Knowledge GPT-4.5 leads
GPT-4.5: 32.5 (#211), Mistral Large: 30.1 (#230)
| Benchmark | GPT-4.5 | Mistral Large |
|---|---|---|
| GPQA Diamond | 68.7% | 51.3% |
| Confabulations | 13.6% | 21.4% |
| LMArena Expert | 1394 | 1232 |
| Humanity's Last Exam | 5.4% | — |
| MMLU-Pro | — | 59.9% |
| Vectara Hallucination Rate | — | 4.5% |
| GPQA (HELM) | — | 43.5% |
| MMLU | — | 80% |
Multimodal Not comparable
GPT-4.5: 37.6 (#71), Mistral Large: —
| Benchmark | GPT-4.5 | Mistral Large |
|---|---|---|
| LMArena Vision | 1195 | — |
| VPCT | 45% | — |
Multilingual GPT-4.5 leads
GPT-4.5: 52.5 (#83), Mistral Large: 40.0 (#219)
| Benchmark | GPT-4.5 | Mistral Large |
|---|---|---|
| LMArena Non-English | 1413 | 1237 |
| LMArena Chinese | 1421 | 1240 |
| LMArena French | 1418 | 1325 |
| LMArena German | 1457 | 1254 |
| LMArena Japanese | 1416 | 1188 |
| LMArena Korean | 1392 | 1202 |
| LMArena Russian | 1419 | 1257 |
| LMArena Spanish | — | 1268 |
Instruction Following GPT-4.5 leads
GPT-4.5: 72.6 (#134), Mistral Large: 67.9 (#191)
| Benchmark | GPT-4.5 | Mistral Large |
|---|---|---|
| LiveBench Instruction Following | 72.3% | 67.9% |
| LMArena Instruction Following | 1404 | 1249 |
| IFEval | — | 87.7% |
Long Context GPT-4.5 leads
GPT-4.5: 40.4 (#155), Mistral Large: 38.3 (#199)
| Benchmark | GPT-4.5 | Mistral Large |
|---|---|---|
| LMArena Longer Query | 1406 | 1261 |
| Fiction.LiveBench | 63.9% | — |
Writing & Preference GPT-4.5 leads
GPT-4.5: 56.9 (#134), Mistral Large: 40.7 (#242)
| Benchmark | GPT-4.5 | Mistral Large |
|---|---|---|
| LMArena Text | 1417 | 1266 |
| LMArena Creative Writing | 1394 | 1243 |
| Short-Story Creative Writing | 75.6% | 69% |
| EQ-Bench Creative Writing | 1258 | 985 |
| LMArena Multi-Turn | 1444 | 1260 |
| LiveBench Language | 61.5% | 39.4% |
| WildBench | — | 80.1% |
Frequently asked questions
Is GPT-4.5 better than Mistral Large?
GPT-4.5 is the stronger model overall, scoring 37.2 to 31.9 on the Noometry Index.
Is GPT-4.5 or Mistral Large better for coding?
GPT-4.5 scores higher on coding benchmarks: 42.2 versus 34.3 in the Noometry coding category.
How many benchmarks do GPT-4.5 and Mistral Large share?
32 benchmarks have published results for both models. GPT-4.5 has 42 scored results on Noometry and Mistral Large has 51.