Model comparison
Codestral vs GPT-5.6 Luna
GPT-5.6 Luna is the stronger model overall, scoring 54.6 to 30.6 on the Noometry Index.
Last verified . 2 shared benchmarks.
Mistral AISummary
- They share 2 benchmarks with published results for both. Codestral scores higher in 0 categories and GPT-5.6 Luna in 2 categories; 2 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-5.6 Luna leads 47.6 to 19.8.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 32.5% for Codestral and 49.1% for GPT-5.6 Luna.
- Both cost about the same: $0.30 input and $0.90 output per million tokens.
- GPT-5.6 Luna accepts more context: 1.05M tokens versus 256K.
Side by side
| Codestral | GPT-5.6 Luna | |
|---|---|---|
| Provider | Mistral AI | OpenAI |
| Noometry Index | 30.6 | 54.6 |
| Released | 2024-05-29 | 2026-07-09 |
| Weights | Proprietary | Proprietary |
| Context window | 256K | 1.05M |
| Max output | 8K | 128K |
| Input $ / M tokens | $0.30 | $0.20 |
| Output $ / M tokens | $0.90 | $1.20 |
| Results tracked | 7 | 52 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.6 Luna leads
Codestral: 27.3 (#321), GPT-5.6 Luna: 54.5 (#28)
| Benchmark | Codestral | GPT-5.6 Luna |
|---|---|---|
| ALE-Bench | 137.78 | 1,667 |
| DeepSWE | — | 67.2% |
| FrontierCode | — | 39.8% |
| Aider Polyglot | 11.1% | — |
| CursorBench | — | 35.9% |
| LMArena WebDev | — | 1519 |
| SciCode | — | 53.6% |
| WeirdML | — | 60.9% |
| BigCodeBench Instruct | 41.8% | — |
| LMArena Coding | — | 1466 |
| BigCodeBench Complete | 52.5% | — |
| HumanEval+ | 73.8% | — |
| MBPP+ | 61.9% | — |
Agentic & Tool Use Not comparable
Codestral: —, GPT-5.6 Luna: 34.4 (#45)
| Benchmark | Codestral | GPT-5.6 Luna |
|---|---|---|
| APEX-Agents | — | 43% |
| BALROG | — | 45.6% |
| GDP.pdf | — | 22.7% |
| Vending-Bench 2 | — | 4,095 |
Reasoning GPT-5.6 Luna leads
Codestral: 19.8 (#251), GPT-5.6 Luna: 47.6 (#43)
| Benchmark | Codestral | GPT-5.6 Luna |
|---|---|---|
| Kagi LLM Benchmark | 32.5% | 49.1% |
| ARC-AGI-2 | — | 59.5% |
| SimpleBench | — | 46.8% |
| NYT Connections (extended) | — | 69.4% |
| ARC-AGI-1 | — | 88% |
| CritPt | — | 20.6% |
| Chess Puzzles | — | 40% |
| LMArena Hard Prompts | — | 1451 |
| Mystery Game Puzzles | — | 21% |
| DTBench | — | 89.1% |
| LMCA | — | 48.5% |
| Surface Evolver Bench | — | 61.9% |
| Epoch Capabilities Index | — | 156.39 |
Math Not comparable
Codestral: —, GPT-5.6 Luna: 77.7 (#14)
| Benchmark | Codestral | GPT-5.6 Luna |
|---|---|---|
| FrontierMath (Tiers 1-3) | — | 82.1% |
| FrontierMath Tier 4 | — | 61% |
| OTIS Mock AIME 2024-2025 | — | 98.3% |
| ProofBench | — | 60% |
| LMArena Math | — | 1458 |
Knowledge Not comparable
Codestral: —, GPT-5.6 Luna: 58.5 (#34)
| Benchmark | Codestral | GPT-5.6 Luna |
|---|---|---|
| GPQA Diamond | — | 91.6% |
| SimpleQA Verified | — | 41% |
| LMArena Expert | — | 1478 |
Multimodal Not comparable
Codestral: —, GPT-5.6 Luna: 42.7 (#28)
| Benchmark | Codestral | GPT-5.6 Luna |
|---|---|---|
| LMArena Vision | — | 1258 |
| Blueprint-Bench 2 | — | 22.6% |
| Furniture Assembly | — | 42.5% |
| LMArena Document | — | 1457 |
Multilingual Not comparable
Codestral: —, GPT-5.6 Luna: 52.8 (#78)
| Benchmark | Codestral | GPT-5.6 Luna |
|---|---|---|
| LMArena Non-English | — | 1417 |
| LMArena Chinese | — | 1470 |
| LMArena French | — | 1456 |
| LMArena German | — | 1454 |
| LMArena Japanese | — | 1411 |
| LMArena Korean | — | 1415 |
| LMArena Russian | — | 1428 |
| LMArena Spanish | — | 1448 |
Instruction Following Not comparable
Codestral: —, GPT-5.6 Luna: 75.6 (#57)
| Benchmark | Codestral | GPT-5.6 Luna |
|---|---|---|
| LMArena Instruction Following | — | 1437 |
Long Context Not comparable
Codestral: —, GPT-5.6 Luna: 43.9 (#82)
| Benchmark | Codestral | GPT-5.6 Luna |
|---|---|---|
| LMArena Longer Query | — | 1436 |
Writing & Preference Not comparable
Codestral: —, GPT-5.6 Luna: 68.0 (#29)
| Benchmark | Codestral | GPT-5.6 Luna |
|---|---|---|
| LMArena Text | — | 1431 |
| LMArena Creative Writing | — | 1396 |
| EQ-Bench Creative Writing | — | 1829 |
| EQ-Bench 4 | — | 1156 |
| LMArena Multi-Turn | — | 1434 |
Frequently asked questions
Is Codestral better than GPT-5.6 Luna?
GPT-5.6 Luna is the stronger model overall, scoring 54.6 to 30.6 on the Noometry Index.
Which is cheaper, Codestral or GPT-5.6 Luna?
Codestral is cheaper. It lists at $0.30 per million input tokens and $0.90 per million output tokens; GPT-5.6 Luna lists at $0.20 and $1.20.
Is Codestral or GPT-5.6 Luna better for coding?
GPT-5.6 Luna scores higher on coding benchmarks: 54.5 versus 27.3 in the Noometry coding category.
Which has the bigger context window?
GPT-5.6 Luna does, with 1.05M tokens against 256K.
How many benchmarks do Codestral and GPT-5.6 Luna share?
2 benchmarks have published results for both models. Codestral has 7 scored results on Noometry and GPT-5.6 Luna has 52.