Model comparison
GPT-6 Astra vs Mistral Medium 3.5
GPT-6 Astra is the stronger model overall, scoring 70.3 to 39.6 on the Noometry Index. Mistral Medium 3.5 costs 6.7× less per token, which makes it the better buy when GPT-6 Astra's lead doesn't matter for your workload.
Last verified . 18 shared benchmarks.
Mistral AISummary
- They share 18 benchmarks with published results for both. GPT-6 Astra scores higher in 9 categories and Mistral Medium 3.5 in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-6 Astra leads 85.1 to 16.6.
- The biggest single-benchmark swing is NYT Connections (extended): 98.1% for GPT-6 Astra and 12.9% for Mistral Medium 3.5.
- Mistral Medium 3.5 is cheaper at $1.50 / $7.50 per million input/output tokens, against $10 / $50 for GPT-6 Astra.
- GPT-6 Astra accepts more context: 1.05M tokens versus 262K.
- Mistral Medium 3.5 has downloadable open weights; the other is API-only.
Side by side
| GPT-6 Astra | Mistral Medium 3.5 | |
|---|---|---|
| Provider | OpenAI | Mistral AI |
| Noometry Index | 70.3 | 39.6 |
| Released | 2026-09-03 | — |
| Weights | Proprietary | Open |
| Context window | 1.05M | 262K |
| Max output | 128K | 210K |
| Input $ / M tokens | $10 | $1.50 |
| Output $ / M tokens | $50 | $7.50 |
| Results tracked | 51 | 22 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-6 Astra leads
GPT-6 Astra: 70.5 (#3), Mistral Medium 3.5: 35.7 (#208)
| Benchmark | GPT-6 Astra | Mistral Medium 3.5 |
|---|---|---|
| LMArena WebDev | 1792 | 1263 |
| LMArena Coding | 1488 | 1461 |
| DeepSWE | 74.1% | — |
| FrontierCode | 53.3% | — |
| FrontierSWE | 65.5% | — |
| SciCode | 56.5% | — |
| WeirdML | 93.6% | — |
| MirrorCode | 46.7% | — |
| ALE-Bench | 2,951 | — |
Agentic & Tool Use Not comparable
GPT-6 Astra: 52.3 (#3), Mistral Medium 3.5: —
| Benchmark | GPT-6 Astra | Mistral Medium 3.5 |
|---|---|---|
| APEX-Agents | 64.7% | — |
| Remote Labor Index | 20.8% | — |
| BALROG | 68.3% | — |
| Vending-Bench 2 | 15,515 | — |
Reasoning GPT-6 Astra leads
GPT-6 Astra: 85.1 (#1), Mistral Medium 3.5: 16.6 (#290)
| Benchmark | GPT-6 Astra | Mistral Medium 3.5 |
|---|---|---|
| NYT Connections (extended) | 98.1% | 12.9% |
| LMArena Hard Prompts | 1464 | 1436 |
| Epoch Capabilities Index | 166.6 | 141.42 |
| ARC-AGI-2 | 95% | — |
| Kagi LLM Benchmark | — | 41.4% |
| ARC-AGI-1 | 98.5% | — |
| CritPt | 31.7% | — |
| Chess Puzzles | 72% | — |
| EBR-Bench | 76.2% | — |
| Mystery Game Puzzles | 84% | — |
| DTBench | 97.3% | — |
| LMCA | 64.4% | — |
| Bench to the Future 3 | 0.14 | — |
Math GPT-6 Astra leads
GPT-6 Astra: 93.4 (#1), Mistral Medium 3.5: 36.4 (#113)
| Benchmark | GPT-6 Astra | Mistral Medium 3.5 |
|---|---|---|
| LMArena Math | 1463 | 1432 |
| FrontierMath (Tiers 1-3) | 93.7% | — |
| FrontierMath Tier 4 | 97.6% | — |
| OTIS Mock AIME 2024-2025 | 100% | — |
| ProofBench | 99% | — |
| FrontierMath Erdős | 2.9% | — |
Knowledge GPT-6 Astra leads
GPT-6 Astra: 75.3 (#1), Mistral Medium 3.5: 39.3 (#123)
| Benchmark | GPT-6 Astra | Mistral Medium 3.5 |
|---|---|---|
| LMArena Expert | 1486 | 1431 |
| GPQA Diamond | 95.8% | — |
| Humanity's Last Exam | 54.8% | — |
| SimpleQA Verified | 75.6% | — |
| Vectara Hallucination Rate | 8.7% | — |
Multimodal GPT-6 Astra leads
GPT-6 Astra: 55.1 (#3), Mistral Medium 3.5: 37.6 (#64)
| Benchmark | GPT-6 Astra | Mistral Medium 3.5 |
|---|---|---|
| LMArena Vision | 1280 | 1222 |
| Blueprint-Bench 2 | 49.7% | — |
| Furniture Assembly | 80% | — |
| LMArena Document | 1468 | — |
Multilingual GPT-6 Astra leads
GPT-6 Astra: 54.0 (#53), Mistral Medium 3.5: 52.1 (#93)
| Benchmark | GPT-6 Astra | Mistral Medium 3.5 |
|---|---|---|
| LMArena Non-English | 1431 | 1405 |
| LMArena Chinese | 1495 | 1443 |
| LMArena French | 1467 | 1448 |
| LMArena Russian | 1434 | 1396 |
| LMArena Spanish | 1412 | 1406 |
| LMArena German | — | 1453 |
| LMArena Korean | — | 1387 |
Instruction Following GPT-6 Astra leads
GPT-6 Astra: 76.6 (#37), Mistral Medium 3.5: 74.7 (#87)
| Benchmark | GPT-6 Astra | Mistral Medium 3.5 |
|---|---|---|
| LMArena Instruction Following | 1454 | 1414 |
Long Context GPT-6 Astra leads
GPT-6 Astra: 44.7 (#57), Mistral Medium 3.5: 43.2 (#101)
| Benchmark | GPT-6 Astra | Mistral Medium 3.5 |
|---|---|---|
| LMArena Longer Query | 1460 | 1414 |
Writing & Preference GPT-6 Astra leads
GPT-6 Astra: 75.7 (#7), Mistral Medium 3.5: 58.6 (#113)
| Benchmark | GPT-6 Astra | Mistral Medium 3.5 |
|---|---|---|
| LMArena Text | 1443 | 1421 |
| LMArena Creative Writing | 1423 | 1372 |
| LMArena Multi-Turn | 1457 | 1423 |
| EQ-Bench Creative Writing | 2173 | — |
| EQ-Bench 4 | — | 993 |
Frequently asked questions
Is GPT-6 Astra better than Mistral Medium 3.5?
GPT-6 Astra is the stronger model overall, scoring 70.3 to 39.6 on the Noometry Index. Mistral Medium 3.5 costs 6.7× less per token, which makes it the better buy when GPT-6 Astra's lead doesn't matter for your workload.
Which is cheaper, GPT-6 Astra or Mistral Medium 3.5?
Mistral Medium 3.5 is cheaper. It lists at $1.50 per million input tokens and $7.50 per million output tokens; GPT-6 Astra lists at $10 and $50.
Is GPT-6 Astra or Mistral Medium 3.5 better for coding?
GPT-6 Astra scores higher on coding benchmarks: 70.5 versus 35.7 in the Noometry coding category.
Which has the bigger context window?
GPT-6 Astra does, with 1.05M tokens against 262K.
How many benchmarks do GPT-6 Astra and Mistral Medium 3.5 share?
18 benchmarks have published results for both models. GPT-6 Astra has 51 scored results on Noometry and Mistral Medium 3.5 has 22.