Model comparison
GPT-3.5-turbo vs Mistral Large
Mistral Large is the stronger model overall, scoring 31.6 to 23.1 on the Noometry Index. GPT-3.5-turbo costs 4.0× less per token, which makes it the better buy when Mistral Large's lead doesn't matter for your workload.
Last verified . 30 shared benchmarks.
Mistral AISummary
- They share 30 benchmarks with published results for both. GPT-3.5-turbo scores higher in 0 categories and Mistral Large in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Mistral Large leads 30.0 to 10.0.
- The biggest single-benchmark swing is MATH Level 5: 15.9% for GPT-3.5-turbo and 50.3% for Mistral Large.
- GPT-3.5-turbo is cheaper at $0.50 / $1.50 per million input/output tokens, against $2 / $6 for Mistral Large.
- Mistral Large accepts more context: 131K tokens versus 16K.
- Mistral Large has downloadable open weights; the other is API-only.
Side by side
| GPT-3.5-turbo | Mistral Large | |
|---|---|---|
| Provider | OpenAI | Mistral AI |
| Noometry Index | 23.1 | 31.6 |
| Released | 2023-03-01 | 2024-02-26 |
| Weights | Proprietary | Open |
| Context window | 16K | 131K |
| Max output | 4K | 16K |
| Input $ / M tokens | $0.50 | $2 |
| Output $ / M tokens | $1.50 | $6 |
| Results tracked | 44 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Mistral Large leads
GPT-3.5-turbo: 23.5 (#326), Mistral Large: 33.8 (#240)
| Benchmark | GPT-3.5-turbo | Mistral Large |
|---|---|---|
| BigCodeBench Instruct | 39.1% | 30% |
| LMArena Coding | 1137 | 1277 |
| BigCodeBench Complete | 50.6% | 38.3% |
| HumanEval+ | 70.7% | 62.2% |
| MBPP+ | 69.7% | 59.5% |
| SciCode | — | 36.2% |
| WeirdML | 3.5% | — |
| LiveBench Coding | — | 47.1% |
| ALE-Bench | — | 264.7 |
Agentic & Tool Use Not comparable
GPT-3.5-turbo: —, Mistral Large: 28.4 (#83)
| Benchmark | GPT-3.5-turbo | Mistral Large |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 38.4% |
| METR Time Horizons | 21.5% | — |
Reasoning Mistral Large leads
GPT-3.5-turbo: 13.3 (#328), Mistral Large: 15.4 (#305)
| Benchmark | GPT-3.5-turbo | Mistral Large |
|---|---|---|
| LMArena Hard Prompts | 1108 | 1257 |
| DTBench | 48.5% | 65.1% |
| LMCA | 9.7% | 16.7% |
| Epoch Capabilities Index | 118.48 | 128.52 |
| ForecastBench | 50.4 | 57.1 |
| SimpleBench | — | 22.5% |
| CritPt | — | 0% |
| Chess Puzzles | 0% | — |
| LiveBench Reasoning | — | 43.5% |
| Mystery Game Puzzles | 3% | — |
| LiveBench Data Analysis | — | 50.1% |
| Adversarial NLI | 58.1% | — |
| BIG-Bench Hard | 61.6% | — |
| CommonsenseQA 2.0 | 57% | — |
| LiveBench | — | 48.4% |
| WinoGrande | 81.6% | — |
Math Mistral Large leads
GPT-3.5-turbo: 6.0 (#321), Mistral Large: 17.1 (#285)
| Benchmark | GPT-3.5-turbo | Mistral Large |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 2.2% | 8.5% |
| LMArena Math | 1142 | 1262 |
| MATH Level 5 | 15.9% | 50.3% |
| FrontierMath (Tiers 1-3) | 0% | — |
| Omni-MATH | — | 28.1% |
| LiveBench Math | — | 42.5% |
| FrontierMath (Feb 2025 set) | — | 0.3% |
| GSM8K | 57.8% | — |
Knowledge Mistral Large leads
GPT-3.5-turbo: 10.0 (#298), Mistral Large: 30.0 (#224)
| Benchmark | GPT-3.5-turbo | Mistral Large |
|---|---|---|
| GPQA Diamond | 28% | 51.3% |
| LMArena Expert | 1070 | 1232 |
| MMLU | 71.4% | 80% |
| MMLU-Pro | — | 59.9% |
| Confabulations | — | 21.4% |
| Vectara Hallucination Rate | — | 4.5% |
| GPQA (HELM) | — | 43.5% |
| ARC (AI2) Challenge | 87.4% | — |
| BoolQ | 87% | — |
| OpenBookQA | 86% | — |
| TriviaQA | 85.8% | — |
Multilingual Mistral Large leads
GPT-3.5-turbo: 31.6 (#253), Mistral Large: 40.1 (#214)
| Benchmark | GPT-3.5-turbo | Mistral Large |
|---|---|---|
| LMArena Non-English | 1108 | 1237 |
| LMArena Chinese | 1075 | 1240 |
| LMArena French | 1117 | 1324 |
| LMArena German | 1090 | 1254 |
| LMArena Japanese | 1043 | 1187 |
| LMArena Korean | 1019 | 1202 |
| LMArena Russian | 1123 | 1257 |
| LMArena Spanish | 1120 | 1268 |
Instruction Following Mistral Large leads
GPT-3.5-turbo: 58.1 (#257), Mistral Large: 68.1 (#186)
| Benchmark | GPT-3.5-turbo | Mistral Large |
|---|---|---|
| LMArena Instruction Following | 1119 | 1249 |
| LiveBench Instruction Following | — | 67.9% |
| IFEval | — | 87.7% |
Long Context Mistral Large leads
GPT-3.5-turbo: 34.0 (#249), Mistral Large: 38.3 (#194)
| Benchmark | GPT-3.5-turbo | Mistral Large |
|---|---|---|
| LMArena Longer Query | 1121 | 1262 |
Writing & Preference Mistral Large leads
GPT-3.5-turbo: 25.4 (#300), Mistral Large: 40.9 (#237)
| Benchmark | GPT-3.5-turbo | Mistral Large |
|---|---|---|
| LMArena Text | 1125 | 1266 |
| LMArena Creative Writing | 1092 | 1243 |
| EQ-Bench Creative Writing | 451 | 985 |
| LMArena Multi-Turn | 1117 | 1260 |
| Short-Story Creative Writing | — | 69% |
| WildBench | — | 80.1% |
| LiveBench Language | — | 39.4% |
Frequently asked questions
Is GPT-3.5-turbo better than Mistral Large?
Mistral Large is the stronger model overall, scoring 31.6 to 23.1 on the Noometry Index. GPT-3.5-turbo costs 4.0× less per token, which makes it the better buy when Mistral Large's lead doesn't matter for your workload.
Which is cheaper, GPT-3.5-turbo or Mistral Large?
GPT-3.5-turbo is cheaper. It lists at $0.50 per million input tokens and $1.50 per million output tokens; Mistral Large lists at $2 and $6.
Is GPT-3.5-turbo or Mistral Large better for coding?
Mistral Large scores higher on coding benchmarks: 33.8 versus 23.5 in the Noometry coding category.
Which has the bigger context window?
Mistral Large does, with 131K tokens against 16K.
How many benchmarks do GPT-3.5-turbo and Mistral Large share?
30 benchmarks have published results for both models. GPT-3.5-turbo has 44 scored results on Noometry and Mistral Large has 51.