Model comparison
GPT-5.4 nano vs Magistral Medium
GPT-5.4 nano is the stronger model overall, scoring 41.9 to 35.2 on the Noometry Index.
Last verified . 22 shared benchmarks.
Mistral AISummary
- They share 22 benchmarks with published results for both. GPT-5.4 nano scores higher in 8 categories and Magistral Medium in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-5.4 nano leads 23.7 to 8.6.
- The biggest single-benchmark swing is ARC-AGI-1: 51.5% for GPT-5.4 nano and 6.1% for Magistral Medium.
- GPT-5.4 nano is cheaper at $0.20 / $1.25 per million input/output tokens, against $2 / $5 for Magistral Medium.
- GPT-5.4 nano accepts more context: 400K tokens versus 262K.
- Magistral Medium has downloadable open weights; the other is API-only.
Side by side
| GPT-5.4 nano | Magistral Medium | |
|---|---|---|
| Provider | OpenAI | Mistral AI |
| Noometry Index | 41.9 | 35.2 |
| Released | 2026-03-17 | 2025-03-17 |
| Weights | Proprietary | Open |
| Context window | 400K | 262K |
| Max output | 128K | 16K |
| Input $ / M tokens | $0.20 | $2 |
| Output $ / M tokens | $1.25 | $5 |
| Results tracked | 40 | 22 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.4 nano leads
GPT-5.4 nano: 43.6 (#84), Magistral Medium: 39.1 (#161)
| Benchmark | GPT-5.4 nano | Magistral Medium |
|---|---|---|
| SciCode | 46.9% | 39.2% |
| LMArena Coding | 1405 | 1319 |
| WeirdML | 49.2% | — |
| ALE-Bench | 1,005 | — |
Reasoning GPT-5.4 nano leads
GPT-5.4 nano: 23.7 (#173), Magistral Medium: 8.6 (#348)
| Benchmark | GPT-5.4 nano | Magistral Medium |
|---|---|---|
| ARC-AGI-2 | 5.7% | 0% |
| Kagi LLM Benchmark | 39.7% | 16.2% |
| ARC-AGI-1 | 51.5% | 6.1% |
| CritPt | 9.3% | 0.3% |
| LMArena Hard Prompts | 1381 | 1267 |
| Chess Puzzles | 30% | — |
| Mystery Game Puzzles | 9% | — |
| DTBench | 80.3% | — |
| LMCA | 36.9% | — |
| Epoch Capabilities Index | 145.81 | — |
| ForecastBench | 57.3 | — |
Math GPT-5.4 nano leads
GPT-5.4 nano: 40.9 (#88), Magistral Medium: 35.1 (#189)
| Benchmark | GPT-5.4 nano | Magistral Medium |
|---|---|---|
| LMArena Math | 1406 | 1250 |
| FrontierMath (Tiers 1-3) | 44.9% | — |
| FrontierMath Tier 4 | 12.2% | — |
| OTIS Mock AIME 2024-2025 | 87.8% | — |
| ProofBench | 5% | — |
| FrontierMath (Feb 2025 set) | 25.9% | — |
| FrontierMath Tier 4 (v1) | 6.3% | — |
Knowledge GPT-5.4 nano leads
GPT-5.4 nano: 41.9 (#103), Magistral Medium: 33.5 (#202)
| Benchmark | GPT-5.4 nano | Magistral Medium |
|---|---|---|
| LMArena Expert | 1396 | 1223 |
| GPQA Diamond | 78.5% | — |
| SimpleQA Verified | 11.7% | — |
| Vectara Hallucination Rate | 3.1% | — |
Multimodal Not comparable
GPT-5.4 nano: 36.7 (#78), Magistral Medium: —
| Benchmark | GPT-5.4 nano | Magistral Medium |
|---|---|---|
| LMArena Vision | 1196 | — |
Multilingual GPT-5.4 nano leads
GPT-5.4 nano: 48.6 (#140), Magistral Medium: 39.6 (#224)
| Benchmark | GPT-5.4 nano | Magistral Medium |
|---|---|---|
| LMArena Non-English | 1359 | 1232 |
| LMArena Chinese | 1392 | 1227 |
| LMArena French | 1396 | 1267 |
| LMArena German | 1367 | 1248 |
| LMArena Japanese | 1343 | 1175 |
| LMArena Korean | 1320 | 1125 |
| LMArena Russian | 1363 | 1224 |
| LMArena Spanish | 1371 | 1271 |
Instruction Following GPT-5.4 nano leads
GPT-5.4 nano: 71.9 (#144), Magistral Medium: 66.0 (#211)
| Benchmark | GPT-5.4 nano | Magistral Medium |
|---|---|---|
| LMArena Instruction Following | 1362 | 1254 |
Long Context GPT-5.4 nano leads
GPT-5.4 nano: 41.6 (#137), Magistral Medium: 39.3 (#183)
| Benchmark | GPT-5.4 nano | Magistral Medium |
|---|---|---|
| LMArena Longer Query | 1366 | 1295 |
Writing & Preference GPT-5.4 nano leads
GPT-5.4 nano: 55.7 (#142), Magistral Medium: 46.3 (#219)
| Benchmark | GPT-5.4 nano | Magistral Medium |
|---|---|---|
| LMArena Text | 1372 | 1255 |
| LMArena Creative Writing | 1314 | 1245 |
| LMArena Multi-Turn | 1382 | 1275 |
Frequently asked questions
Is GPT-5.4 nano better than Magistral Medium?
GPT-5.4 nano is the stronger model overall, scoring 41.9 to 35.2 on the Noometry Index.
Which is cheaper, GPT-5.4 nano or Magistral Medium?
GPT-5.4 nano is cheaper. It lists at $0.20 per million input tokens and $1.25 per million output tokens; Magistral Medium lists at $2 and $5.
Is GPT-5.4 nano or Magistral Medium better for coding?
GPT-5.4 nano scores higher on coding benchmarks: 43.6 versus 39.1 in the Noometry coding category.
Which has the bigger context window?
GPT-5.4 nano does, with 400K tokens against 262K.
How many benchmarks do GPT-5.4 nano and Magistral Medium share?
22 benchmarks have published results for both models. GPT-5.4 nano has 40 scored results on Noometry and Magistral Medium has 22.