Model comparison
GPT-5.4 mini vs Mistral Small
GPT-5.4 mini is the stronger model overall, scoring 45.0 to 33.4 on the Noometry Index. Mistral Small costs 6.4× less per token, which makes it the better buy when GPT-5.4 mini'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-5.4 mini scores higher in 10 categories and Mistral Small in 0 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.4 mini leads 45.5 to 16.4.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 88.9% for GPT-5.4 mini and 5.8% for Mistral Small.
- Mistral Small is cheaper at $0.15 / $0.60 per million input/output tokens, against $0.75 / $4.50 for GPT-5.4 mini.
- GPT-5.4 mini accepts more context: 400K tokens versus 262K.
- Mistral Small has downloadable open weights; the other is API-only.
Side by side
| GPT-5.4 mini | Mistral Small | |
|---|---|---|
| Provider | OpenAI | Mistral AI |
| Noometry Index | 45.0 | 33.4 |
| Released | 2026-03-17 | 2024-02-26 |
| Weights | Proprietary | Open |
| Context window | 400K | 262K |
| Max output | 128K | 256K |
| Input $ / M tokens | $0.75 | $0.15 |
| Output $ / M tokens | $4.50 | $0.60 |
| Results tracked | 46 | 39 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.4 mini leads
GPT-5.4 mini: 45.2 (#72), Mistral Small: 34.0 (#247)
| Benchmark | GPT-5.4 mini | Mistral Small |
|---|---|---|
| SciCode | 49.9% | 26.5% |
| LMArena Coding | 1438 | 1362 |
| ALE-Bench | 1,189 | 497.62 |
| FrontierCode | 27% | — |
| LMArena WebDev | 1397 | — |
| WeirdML | 60.3% | — |
| BigCodeBench Instruct | — | 36.1% |
| LiveBench Coding | — | 36.2% |
| BigCodeBench Complete | — | 46.6% |
Agentic & Tool Use GPT-5.4 mini leads
GPT-5.4 mini: 29.9 (#81), Mistral Small: 28.1 (#93)
| Benchmark | GPT-5.4 mini | Mistral Small |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 37.1% |
| DeepResearch Bench | 36.3% | — |
Reasoning GPT-5.4 mini leads
GPT-5.4 mini: 30.4 (#85), Mistral Small: 19.8 (#250)
| Benchmark | GPT-5.4 mini | Mistral Small |
|---|---|---|
| Kagi LLM Benchmark | 37.9% | 37.8% |
| CritPt | 10% | 0% |
| LMArena Hard Prompts | 1424 | 1335 |
| DTBench | 80% | 70.9% |
| LMCA | 40.8% | 20.6% |
| ARC-AGI-2 | 18.9% | — |
| NYT Connections (extended) | 61.8% | — |
| ARC-AGI-1 | 63.7% | — |
| Chess Puzzles | 24% | — |
| Thematic Generalization | 61.7% | — |
| LiveBench Reasoning | — | 44.8% |
| Mystery Game Puzzles | 11% | — |
| LiveBench Data Analysis | — | 53.7% |
| Epoch Capabilities Index | 148.84 | — |
| ForecastBench | 57 | — |
| LiveBench | — | 44% |
Math GPT-5.4 mini leads
GPT-5.4 mini: 45.5 (#75), Mistral Small: 16.4 (#293)
| Benchmark | GPT-5.4 mini | Mistral Small |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 88.9% | 5.8% |
| LMArena Math | 1419 | 1341 |
| FrontierMath (Tiers 1-3) | 51.2% | — |
| FrontierMath Tier 4 | 9.8% | — |
| ProofBench | 21% | — |
| LiveBench Math | — | 39.9% |
| MATH Level 5 | — | 46.8% |
| FrontierMath (Feb 2025 set) | 28.3% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge GPT-5.4 mini leads
GPT-5.4 mini: 51.5 (#67), Mistral Small: 31.0 (#222)
| Benchmark | GPT-5.4 mini | Mistral Small |
|---|---|---|
| GPQA Diamond | 86.9% | 47.5% |
| Vectara Hallucination Rate | 5.5% | 5.1% |
| LMArena Expert | 1435 | 1291 |
| SimpleQA Verified | 29.4% | — |
| MMLU | — | 68.7% |
Multimodal GPT-5.4 mini leads
GPT-5.4 mini: 39.7 (#56), Mistral Small: 33.5 (#96)
| Benchmark | GPT-5.4 mini | Mistral Small |
|---|---|---|
| LMArena Vision | 1245 | 1142 |
Multilingual GPT-5.4 mini leads
GPT-5.4 mini: 51.9 (#96), Mistral Small: 45.5 (#169)
| Benchmark | GPT-5.4 mini | Mistral Small |
|---|---|---|
| LMArena Non-English | 1405 | 1315 |
| LMArena Chinese | 1446 | 1340 |
| LMArena French | 1440 | 1337 |
| LMArena German | 1409 | 1340 |
| LMArena Japanese | 1374 | 1275 |
| LMArena Korean | 1368 | 1259 |
| LMArena Russian | 1417 | 1324 |
| LMArena Spanish | 1405 | 1346 |
Instruction Following GPT-5.4 mini leads
GPT-5.4 mini: 74.1 (#102), Mistral Small: 66.4 (#209)
| Benchmark | GPT-5.4 mini | Mistral Small |
|---|---|---|
| LMArena Instruction Following | 1405 | 1310 |
| LiveBench Instruction Following | — | 63.7% |
Long Context GPT-5.4 mini leads
GPT-5.4 mini: 43.0 (#112), Mistral Small: 40.4 (#156)
| Benchmark | GPT-5.4 mini | Mistral Small |
|---|---|---|
| LMArena Longer Query | 1407 | 1327 |
Writing & Preference GPT-5.4 mini leads
GPT-5.4 mini: 64.0 (#58), Mistral Small: 52.5 (#171)
| Benchmark | GPT-5.4 mini | Mistral Small |
|---|---|---|
| LMArena Text | 1412 | 1338 |
| LMArena Creative Writing | 1370 | 1305 |
| LMArena Multi-Turn | 1429 | 1344 |
| EQ-Bench Creative Writing | 1665 | — |
| LiveBench Language | — | 30.5% |
Frequently asked questions
Is GPT-5.4 mini better than Mistral Small?
GPT-5.4 mini is the stronger model overall, scoring 45.0 to 33.4 on the Noometry Index. Mistral Small costs 6.4× less per token, which makes it the better buy when GPT-5.4 mini's lead doesn't matter for your workload.
Which is cheaper, GPT-5.4 mini or Mistral Small?
Mistral Small is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; GPT-5.4 mini lists at $0.75 and $4.50.
Is GPT-5.4 mini or Mistral Small better for coding?
GPT-5.4 mini scores higher on coding benchmarks: 45.2 versus 34.0 in the Noometry coding category.
Which has the bigger context window?
GPT-5.4 mini does, with 400K tokens against 262K.
How many benchmarks do GPT-5.4 mini and Mistral Small share?
27 benchmarks have published results for both models. GPT-5.4 mini has 46 scored results on Noometry and Mistral Small has 39.