Model comparison
GPT-4 Turbo vs Mistral 7B
GPT-4 Turbo is the stronger model overall, scoring 30.5 to 23.0 on the Noometry Index. Mistral 7B costs 60× less per token, which makes it the better buy when GPT-4 Turbo's lead doesn't matter for your workload.
Last verified . 27 shared benchmarks.
Mistral AISummary
- They share 27 benchmarks with published results for both. GPT-4 Turbo scores higher in 8 categories and Mistral 7B in 0 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where GPT-4 Turbo leads 47.7 to 30.7.
- The biggest single-benchmark swing is MATH Level 5: 46.7% for GPT-4 Turbo and 3.7% for Mistral 7B.
- Mistral 7B is cheaper at $0.25 / $0.25 per million input/output tokens, against $10 / $30 for GPT-4 Turbo.
- GPT-4 Turbo accepts more context: 128K tokens versus 8K.
- Mistral 7B has downloadable open weights; the other is API-only.
Side by side
| GPT-4 Turbo | Mistral 7B | |
|---|---|---|
| Provider | OpenAI | Mistral AI |
| Noometry Index | 30.5 | 23.0 |
| Released | 2023-11-06 | 2023-09-27 |
| Weights | Proprietary | Open |
| Context window | 128K | 8K |
| Max output | 4K | 8K |
| Input $ / M tokens | $10 | $0.25 |
| Output $ / M tokens | $30 | $0.25 |
| Results tracked | 36 | 37 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-4 Turbo leads
GPT-4 Turbo: 33.8 (#249), Mistral 7B: 26.4 (#326)
| Benchmark | GPT-4 Turbo | Mistral 7B |
|---|---|---|
| BigCodeBench Instruct | 48.2% | 19.5% |
| LMArena Coding | 1268 | 1082 |
| BigCodeBench Complete | 58.2% | 27.3% |
| HumanEval+ | 86.6% | 36% |
| MBPP+ | 73.3% | 42.1% |
| WeirdML | 18% | — |
Agentic & Tool Use Not comparable
GPT-4 Turbo: —, Mistral 7B: —
| Benchmark | GPT-4 Turbo | Mistral 7B |
|---|---|---|
| METR Time Horizons | 36.7% | — |
Reasoning GPT-4 Turbo leads
GPT-4 Turbo: 15.3 (#317), Mistral 7B: 13.1 (#336)
| Benchmark | GPT-4 Turbo | Mistral 7B |
|---|---|---|
| Chess Puzzles | 6% | 0% |
| LMArena Hard Prompts | 1251 | 1067 |
| DTBench | 61.6% | 42.5% |
| Epoch Capabilities Index | 127.25 | 112.21 |
| SimpleBench | 25.1% | — |
| LMCA | 9.8% | — |
| Adversarial NLI | — | 47.1% |
| BIG-Bench Hard | — | 56.1% |
| ForecastBench | 59.4 | — |
| HellaSwag | — | 81% |
| PIQA | — | 83% |
| WinoGrande | — | 75.3% |
Math Too close to call
GPT-4 Turbo: 9.0 (#322), Mistral 7B: 8.1 (#325)
| Benchmark | GPT-4 Turbo | Mistral 7B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 6.7% | 0.3% |
| LMArena Math | 1272 | 1085 |
| MATH Level 5 | 46.7% | 3.7% |
| FrontierMath (Tiers 1-3) | 0.7% | — |
| GSM8K | — | 54.4% |
Knowledge GPT-4 Turbo leads
GPT-4 Turbo: 24.3 (#268), Mistral 7B: 7.4 (#311)
| Benchmark | GPT-4 Turbo | Mistral 7B |
|---|---|---|
| GPQA Diamond | 46.6% | 15.2% |
| LMArena Expert | 1223 | 1036 |
| MMLU | 81.3% | 62.5% |
| Confabulations | 28.4% | — |
| ARC (AI2) Challenge | — | 78.6% |
| BoolQ | — | 87.4% |
| OpenBookQA | — | 79.8% |
| TriviaQA | — | 75.2% |
Multimodal Not comparable
GPT-4 Turbo: 30.6 (#110), Mistral 7B: —
| Benchmark | GPT-4 Turbo | Mistral 7B |
|---|---|---|
| LMArena Vision | 1090 | — |
Multilingual GPT-4 Turbo leads
GPT-4 Turbo: 40.5 (#216), Mistral 7B: 25.8 (#283)
| Benchmark | GPT-4 Turbo | Mistral 7B |
|---|---|---|
| LMArena Non-English | 1245 | 1012 |
| LMArena Chinese | 1242 | 1009 |
| LMArena French | 1276 | 1037 |
| LMArena German | 1259 | 987 |
| LMArena Japanese | 1194 | 878 |
| LMArena Russian | 1259 | 1018 |
| LMArena Spanish | 1260 | 1026 |
| LMArena Korean | 1187 | — |
Instruction Following GPT-4 Turbo leads
GPT-4 Turbo: 65.8 (#216), Mistral 7B: 54.2 (#280)
| Benchmark | GPT-4 Turbo | Mistral 7B |
|---|---|---|
| LMArena Instruction Following | 1249 | 1060 |
Long Context GPT-4 Turbo leads
GPT-4 Turbo: 38.0 (#206), Mistral 7B: 32.2 (#271)
| Benchmark | GPT-4 Turbo | Mistral 7B |
|---|---|---|
| LMArena Longer Query | 1254 | 1060 |
Writing & Preference GPT-4 Turbo leads
GPT-4 Turbo: 47.7 (#206), Mistral 7B: 30.7 (#286)
| Benchmark | GPT-4 Turbo | Mistral 7B |
|---|---|---|
| LMArena Text | 1272 | 1090 |
| LMArena Creative Writing | 1269 | 1068 |
| LMArena Multi-Turn | 1267 | 1062 |
Frequently asked questions
Is GPT-4 Turbo better than Mistral 7B?
GPT-4 Turbo is the stronger model overall, scoring 30.5 to 23.0 on the Noometry Index. Mistral 7B costs 60× less per token, which makes it the better buy when GPT-4 Turbo's lead doesn't matter for your workload.
Which is cheaper, GPT-4 Turbo or Mistral 7B?
Mistral 7B is cheaper. It lists at $0.25 per million input tokens and $0.25 per million output tokens; GPT-4 Turbo lists at $10 and $30.
Is GPT-4 Turbo or Mistral 7B better for coding?
GPT-4 Turbo scores higher on coding benchmarks: 33.8 versus 26.4 in the Noometry coding category.
Which has the bigger context window?
GPT-4 Turbo does, with 128K tokens against 8K.
How many benchmarks do GPT-4 Turbo and Mistral 7B share?
27 benchmarks have published results for both models. GPT-4 Turbo has 36 scored results on Noometry and Mistral 7B has 37.