Model comparison
GPT-6 Luna vs Mistral Medium
GPT-6 Luna is the stronger model overall, scoring 53.3 to 36.3 on the Noometry Index.
Last verified . 27 shared benchmarks.
Mistral AISummary
- They share 27 benchmarks with published results for both. GPT-6 Luna scores higher in 8 categories and Mistral Medium in 2 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-6 Luna leads 76.1 to 28.1.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 98.9% for GPT-6 Luna and 32.2% for Mistral Medium.
- GPT-6 Luna is cheaper at $0.10 / $0.50 per million input/output tokens, against $1.50 / $7.50 for Mistral Medium.
- GPT-6 Luna accepts more context: 1.05M tokens versus 262K.
- Mistral Medium has downloadable open weights; the other is API-only.
Side by side
| GPT-6 Luna | Mistral Medium | |
|---|---|---|
| Provider | OpenAI | Mistral AI |
| Noometry Index | 53.3 | 36.3 |
| Released | 2026-09-22 | 2023-12-11 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 262K |
| Max output | 128K | 262K |
| Input $ / M tokens | $0.10 | $1.50 |
| Output $ / M tokens | $0.50 | $7.50 |
| Results tracked | 42 | 36 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-6 Luna leads
GPT-6 Luna: 55.5 (#25), Mistral Medium: 34.2 (#243)
| Benchmark | GPT-6 Luna | Mistral Medium |
|---|---|---|
| FrontierCode | 42.4% | 8% |
| SciCode | 54.6% | 40.2% |
| LMArena Coding | 1439 | 1434 |
| ALE-Bench | 1,577 | 763.98 |
| DeepSWE | 66.6% | — |
| LMArena WebDev | 1581 | — |
| WeirdML | — | 43.7% |
Agentic & Tool Use GPT-6 Luna leads
GPT-6 Luna: 33.3 (#54), Mistral Medium: 28.3 (#90)
| Benchmark | GPT-6 Luna | Mistral Medium |
|---|---|---|
| APEX-Agents | 44.3% | — |
| Berkeley Function Calling Leaderboard | — | 37.7% |
| GDP.pdf | 23% | — |
Reasoning GPT-6 Luna leads
GPT-6 Luna: 48.2 (#41), Mistral Medium: 24.0 (#167)
| Benchmark | GPT-6 Luna | Mistral Medium |
|---|---|---|
| CritPt | 19.4% | 0% |
| LMArena Hard Prompts | 1411 | 1426 |
| DTBench | 90.1% | 75.5% |
| LMCA | 44.5% | 26.1% |
| ARC-AGI-2 | 59.3% | — |
| Kagi LLM Benchmark | — | 50% |
| NYT Connections (extended) | 68.7% | — |
| ARC-AGI-1 | 86.7% | — |
| Chess Puzzles | 31% | — |
| Mystery Game Puzzles | 7% | — |
| Surface Evolver Bench | — | 26.9% |
| Epoch Capabilities Index | 156.28 | — |
Math GPT-6 Luna leads
GPT-6 Luna: 76.1 (#15), Mistral Medium: 28.1 (#245)
| Benchmark | GPT-6 Luna | Mistral Medium |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 98.9% | 32.2% |
| ProofBench | 64% | 9% |
| LMArena Math | 1416 | 1408 |
| FrontierMath (Tiers 1-3) | 78.9% | — |
| FrontierMath Tier 4 | 56.1% | — |
| MATH Level 5 | — | 81.6% |
| FrontierMath (Feb 2025 set) | — | 0.3% |
Knowledge GPT-6 Luna leads
GPT-6 Luna: 57.0 (#41), Mistral Medium: 25.0 (#265)
| Benchmark | GPT-6 Luna | Mistral Medium |
|---|---|---|
| GPQA Diamond | 90.5% | 59.5% |
| LMArena Expert | 1444 | 1408 |
| Humanity's Last Exam | — | 4.5% |
| SimpleQA Verified | 41.4% | — |
| Vectara Hallucination Rate | — | 22.7% |
Multimodal GPT-6 Luna leads
GPT-6 Luna: 42.4 (#30), Mistral Medium: 35.3 (#88)
| Benchmark | GPT-6 Luna | Mistral Medium |
|---|---|---|
| LMArena Vision | 1217 | 1172 |
| Blueprint-Bench 2 | 31.2% | — |
| Furniture Assembly | 44.2% | — |
Multilingual Mistral Medium leads
GPT-6 Luna: 50.5 (#117), Mistral Medium: 52.1 (#91)
| Benchmark | GPT-6 Luna | Mistral Medium |
|---|---|---|
| LMArena Non-English | 1386 | 1408 |
| LMArena Chinese | 1433 | 1447 |
| LMArena French | 1420 | 1459 |
| LMArena German | 1369 | 1432 |
| LMArena Japanese | 1369 | 1378 |
| LMArena Korean | 1360 | 1380 |
| LMArena Russian | 1394 | 1411 |
| LMArena Spanish | 1393 | 1433 |
Instruction Following Too close to call
GPT-6 Luna: 74.3 (#99), Mistral Medium: 73.7 (#116)
| Benchmark | GPT-6 Luna | Mistral Medium |
|---|---|---|
| LMArena Instruction Following | 1409 | 1398 |
Long Context Too close to call
GPT-6 Luna: 43.0 (#111), Mistral Medium: 42.9 (#114)
| Benchmark | GPT-6 Luna | Mistral Medium |
|---|---|---|
| LMArena Longer Query | 1409 | 1406 |
Writing & Preference Mistral Medium leads
GPT-6 Luna: 58.3 (#119), Mistral Medium: 60.0 (#103)
| Benchmark | GPT-6 Luna | Mistral Medium |
|---|---|---|
| LMArena Text | 1391 | 1424 |
| LMArena Creative Writing | 1363 | 1391 |
| LMArena Multi-Turn | 1396 | 1418 |
| Short-Story Creative Writing | — | 77.3% |
Frequently asked questions
Is GPT-6 Luna better than Mistral Medium?
GPT-6 Luna is the stronger model overall, scoring 53.3 to 36.3 on the Noometry Index.
Which is cheaper, GPT-6 Luna or Mistral Medium?
GPT-6 Luna is cheaper. It lists at $0.10 per million input tokens and $0.50 per million output tokens; Mistral Medium lists at $1.50 and $7.50.
Is GPT-6 Luna or Mistral Medium better for coding?
GPT-6 Luna scores higher on coding benchmarks: 55.5 versus 34.2 in the Noometry coding category.
Which has the bigger context window?
GPT-6 Luna does, with 1.05M tokens against 262K.
How many benchmarks do GPT-6 Luna and Mistral Medium share?
27 benchmarks have published results for both models. GPT-6 Luna has 42 scored results on Noometry and Mistral Medium has 36.