Model comparison
GPT-4 Turbo vs Ministral 8B
GPT-4 Turbo is the stronger model overall, scoring 30.5 to 28.2 on the Noometry Index. Ministral 8B costs 100× less per token, which makes it the better buy when GPT-4 Turbo's lead doesn't matter for your workload.
Last verified . 15 shared benchmarks.
Mistral AISummary
- They share 15 benchmarks with published results for both. GPT-4 Turbo scores higher in 5 categories and Ministral 8B in 3 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Ministral 8B leads 25.7 to 9.0.
- The biggest single-benchmark swing is MATH Level 5: 46.7% for GPT-4 Turbo and 14.9% for Ministral 8B.
- Ministral 8B is cheaper at $0.15 / $0.15 per million input/output tokens, against $10 / $30 for GPT-4 Turbo.
- Ministral 8B accepts more context: 262K tokens versus 128K.
- Ministral 8B has downloadable open weights; the other is API-only.
Side by side
| GPT-4 Turbo | Ministral 8B | |
|---|---|---|
| Provider | OpenAI | Mistral AI |
| Noometry Index | 30.5 | 28.2 |
| Released | 2023-11-06 | 2024-10-01 |
| Weights | Proprietary | Open |
| Context window | 128K | 262K |
| Max output | 4K | 262K |
| Input $ / M tokens | $10 | $0.15 |
| Output $ / M tokens | $30 | $0.15 |
| Results tracked | 36 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Ministral 8B leads
GPT-4 Turbo: 33.8 (#249), Ministral 8B: 35.0 (#230)
| Benchmark | GPT-4 Turbo | Ministral 8B |
|---|---|---|
| LMArena Coding | 1268 | 1202 |
| WeirdML | 18% | — |
| BigCodeBench Instruct | 48.2% | — |
| BigCodeBench Complete | 58.2% | — |
| HumanEval+ | 86.6% | — |
| MBPP+ | 73.3% | — |
Agentic & Tool Use Not comparable
GPT-4 Turbo: —, Ministral 8B: 16.4 (#148)
| Benchmark | GPT-4 Turbo | Ministral 8B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 11.1% |
| METR Time Horizons | 36.7% | — |
Reasoning Ministral 8B leads
GPT-4 Turbo: 15.3 (#317), Ministral 8B: 18.4 (#281)
| Benchmark | GPT-4 Turbo | Ministral 8B |
|---|---|---|
| LMArena Hard Prompts | 1251 | 1191 |
| DTBench | 61.6% | 45.7% |
| SimpleBench | 25.1% | — |
| Chess Puzzles | 6% | — |
| LMCA | 9.8% | — |
| Epoch Capabilities Index | 127.25 | — |
| ForecastBench | 59.4 | — |
Math Ministral 8B leads
GPT-4 Turbo: 9.0 (#322), Ministral 8B: 25.7 (#267)
| Benchmark | GPT-4 Turbo | Ministral 8B |
|---|---|---|
| LMArena Math | 1272 | 1188 |
| MATH Level 5 | 46.7% | 14.9% |
| FrontierMath (Tiers 1-3) | 0.7% | — |
| OTIS Mock AIME 2024-2025 | 6.7% | — |
Knowledge GPT-4 Turbo leads
GPT-4 Turbo: 24.3 (#268), Ministral 8B: 12.6 (#297)
| Benchmark | GPT-4 Turbo | Ministral 8B |
|---|---|---|
| GPQA Diamond | 46.6% | 27.1% |
| LMArena Expert | 1223 | 1170 |
| Confabulations | 28.4% | — |
| Vectara Hallucination Rate | — | 7.4% |
| MMLU | 81.3% | — |
Multimodal Not comparable
GPT-4 Turbo: 30.6 (#110), Ministral 8B: —
| Benchmark | GPT-4 Turbo | Ministral 8B |
|---|---|---|
| LMArena Vision | 1090 | — |
Multilingual GPT-4 Turbo leads
GPT-4 Turbo: 40.5 (#216), Ministral 8B: 35.1 (#247)
| Benchmark | GPT-4 Turbo | Ministral 8B |
|---|---|---|
| LMArena Non-English | 1245 | 1165 |
| LMArena Chinese | 1242 | 1193 |
| LMArena Russian | 1259 | 1195 |
| LMArena French | 1276 | — |
| LMArena German | 1259 | — |
| LMArena Japanese | 1194 | — |
| LMArena Korean | 1187 | — |
| LMArena Spanish | 1260 | — |
Instruction Following GPT-4 Turbo leads
GPT-4 Turbo: 65.8 (#216), Ministral 8B: 60.5 (#250)
| Benchmark | GPT-4 Turbo | Ministral 8B |
|---|---|---|
| LMArena Instruction Following | 1249 | 1161 |
Long Context GPT-4 Turbo leads
GPT-4 Turbo: 38.0 (#206), Ministral 8B: 36.7 (#227)
| Benchmark | GPT-4 Turbo | Ministral 8B |
|---|---|---|
| LMArena Longer Query | 1254 | 1212 |
Writing & Preference GPT-4 Turbo leads
GPT-4 Turbo: 47.7 (#206), Ministral 8B: 39.6 (#246)
| Benchmark | GPT-4 Turbo | Ministral 8B |
|---|---|---|
| LMArena Text | 1272 | 1191 |
| LMArena Creative Writing | 1269 | 1175 |
| LMArena Multi-Turn | 1267 | 1166 |
Frequently asked questions
Is GPT-4 Turbo better than Ministral 8B?
GPT-4 Turbo is the stronger model overall, scoring 30.5 to 28.2 on the Noometry Index. Ministral 8B costs 100× 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 Ministral 8B?
Ministral 8B is cheaper. It lists at $0.15 per million input tokens and $0.15 per million output tokens; GPT-4 Turbo lists at $10 and $30.
Is GPT-4 Turbo or Ministral 8B better for coding?
Ministral 8B scores higher on coding benchmarks: 35.0 versus 33.8 in the Noometry coding category.
Which has the bigger context window?
Ministral 8B does, with 262K tokens against 128K.
How many benchmarks do GPT-4 Turbo and Ministral 8B share?
15 benchmarks have published results for both models. GPT-4 Turbo has 36 scored results on Noometry and Ministral 8B has 17.