Model comparison
Devstral Small 2505 vs GPT-6 Luna
GPT-6 Luna is the stronger model overall, scoring 45.1 to 33.9 on the Noometry Index.
Last verified . 0 shared benchmarks.
Mistral AISummary
- The widest gap is in coding, where GPT-6 Luna leads 54.7 to 38.6.
- Devstral Small 2505 is cheaper at $0.10 / $0.30 per million input/output tokens, against $0.10 / $0.50 for GPT-6 Luna.
- GPT-6 Luna accepts more context: 1.05M tokens versus 128K.
- Devstral Small 2505 has downloadable open weights; the other is API-only.
Side by side
| Devstral Small 2505 | GPT-6 Luna | |
|---|---|---|
| Provider | Mistral AI | OpenAI |
| Noometry Index | 33.9 | 45.1 |
| Released | 2025-05-07 | 2026-09-22 |
| Weights | Open | Proprietary |
| Context window | 128K | 1.05M |
| Max output | 128K | 128K |
| Input $ / M tokens | $0.10 | $0.10 |
| Output $ / M tokens | $0.30 | $0.50 |
| Results tracked | 4 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-6 Luna leads
Devstral Small 2505: 38.6 (#160), GPT-6 Luna: 54.7 (#24)
| Benchmark | Devstral Small 2505 | GPT-6 Luna |
|---|---|---|
| DeepSWE | — | 66.6% |
| FrontierCode | — | 42.4% |
| SWE-bench Verified (bash only) | 56.4% | — |
| LMArena WebDev | — | 1583 |
| SciCode | 28.8% | — |
| LMArena Coding | — | 1448 |
Reasoning GPT-6 Luna leads
Devstral Small 2505: 19.1 (#243), GPT-6 Luna: 31.7 (#68)
| Benchmark | Devstral Small 2505 | GPT-6 Luna |
|---|---|---|
| Kagi LLM Benchmark | 37.7% | — |
| NYT Connections (extended) | — | 68.7% |
| CritPt | 0% | — |
| LMArena Hard Prompts | — | 1415 |
Math Not comparable
Devstral Small 2505: —, GPT-6 Luna: 35.6 (#140)
| Benchmark | Devstral Small 2505 | GPT-6 Luna |
|---|---|---|
| LMArena Math | — | 1396 |
Knowledge Not comparable
Devstral Small 2505: —, GPT-6 Luna: 40.1 (#110)
| Benchmark | Devstral Small 2505 | GPT-6 Luna |
|---|---|---|
| LMArena Expert | — | 1458 |
Multimodal Not comparable
Devstral Small 2505: —, GPT-6 Luna: 42.0 (#27)
| Benchmark | Devstral Small 2505 | GPT-6 Luna |
|---|---|---|
| Furniture Assembly | — | 44.2% |
Multilingual Not comparable
Devstral Small 2505: —, GPT-6 Luna: 51.0 (#111)
| Benchmark | Devstral Small 2505 | GPT-6 Luna |
|---|---|---|
| LMArena Non-English | — | 1390 |
| LMArena Chinese | — | 1425 |
| LMArena Russian | — | 1392 |
| LMArena Spanish | — | 1416 |
Instruction Following Not comparable
Devstral Small 2505: —, GPT-6 Luna: 74.7 (#85)
| Benchmark | Devstral Small 2505 | GPT-6 Luna |
|---|---|---|
| LMArena Instruction Following | — | 1414 |
Long Context Not comparable
Devstral Small 2505: —, GPT-6 Luna: 43.3 (#95)
| Benchmark | Devstral Small 2505 | GPT-6 Luna |
|---|---|---|
| LMArena Longer Query | — | 1417 |
Writing & Preference Not comparable
Devstral Small 2505: —, GPT-6 Luna: 58.4 (#115)
| Benchmark | Devstral Small 2505 | GPT-6 Luna |
|---|---|---|
| LMArena Text | — | 1392 |
| LMArena Creative Writing | — | 1357 |
| LMArena Multi-Turn | — | 1400 |
Frequently asked questions
Is Devstral Small 2505 better than GPT-6 Luna?
GPT-6 Luna is the stronger model overall, scoring 45.1 to 33.9 on the Noometry Index.
Which is cheaper, Devstral Small 2505 or GPT-6 Luna?
Devstral Small 2505 is cheaper. It lists at $0.10 per million input tokens and $0.30 per million output tokens; GPT-6 Luna lists at $0.10 and $0.50.
Is Devstral Small 2505 or GPT-6 Luna better for coding?
GPT-6 Luna scores higher on coding benchmarks: 54.7 versus 38.6 in the Noometry coding category.
Which has the bigger context window?
GPT-6 Luna does, with 1.05M tokens against 128K.
How many benchmarks do Devstral Small 2505 and GPT-6 Luna share?
0 benchmarks have published results for both models. Devstral Small 2505 has 4 scored results on Noometry and GPT-6 Luna has 18.