Model comparison
Devstral Small 2505 vs o3
o3 is the stronger model overall, scoring 47.4 to 33.9 on the Noometry Index. Devstral Small 2505 costs 23× less per token, which makes it the better buy when o3's lead doesn't matter for your workload.
Last verified . 3 shared benchmarks.
Mistral AISummary
- They share 3 benchmarks with published results for both. Devstral Small 2505 scores higher in 0 categories and o3 in 2 categories; 2 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where o3 leads 32.1 to 19.1.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 37.7% for Devstral Small 2505 and 67.6% for o3.
- Devstral Small 2505 is cheaper at $0.10 / $0.30 per million input/output tokens, against $2 / $8 for o3.
- o3 accepts more context: 200K tokens versus 128K.
- Devstral Small 2505 has downloadable open weights; the other is API-only.
Side by side
| Devstral Small 2505 | o3 | |
|---|---|---|
| Provider | Mistral AI | OpenAI |
| Noometry Index | 33.9 | 47.4 |
| Released | 2025-05-07 | 2025-04-16 |
| Weights | Open | Proprietary |
| Context window | 128K | 200K |
| Max output | 128K | 100K |
| Input $ / M tokens | $0.10 | $2 |
| Output $ / M tokens | $0.30 | $8 |
| Results tracked | 4 | 63 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o3 leads
Devstral Small 2505: 38.6 (#160), o3: 47.3 (#57)
| Benchmark | Devstral Small 2505 | o3 |
|---|---|---|
| SWE-bench Verified (bash only) | 56.4% | 58.4% |
| SWE-bench Verified | — | 62.3% |
| Aider Polyglot | — | 81.3% |
| SciCode | 28.8% | — |
| GSO | — | 8.8% |
| WeirdML | — | 52.4% |
| LMArena Coding | — | 1408 |
| CadEval | — | 74% |
| ALE-Bench | — | 933.55 |
Agentic & Tool Use Not comparable
Devstral Small 2505: —, o3: 34.3 (#42)
| Benchmark | Devstral Small 2505 | o3 |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 63% |
| GDPval | — | 30.8% |
| DeepResearch Bench | — | 45.2% |
| OSWorld | — | 23% |
| LMArena Search | — | 1144 |
| METR Time Horizons | — | 65.4% |
Reasoning o3 leads
Devstral Small 2505: 19.1 (#243), o3: 32.1 (#66)
| Benchmark | Devstral Small 2505 | o3 |
|---|---|---|
| Kagi LLM Benchmark | 37.7% | 67.6% |
| CritPt | 0% | 1.4% |
| ARC-AGI-2 | — | 6.5% |
| SimpleBench | — | 53.1% |
| ARC-AGI-1 | — | 60.8% |
| Chess Puzzles | — | 38% |
| EnigmaEval | — | 13.1% |
| LMArena Hard Prompts | — | 1402 |
| Mystery Game Puzzles | — | 29% |
| DTBench | — | 84.8% |
| LMCA | — | 39.7% |
| Epoch Capabilities Index | — | 146.88 |
| ForecastBench | — | 62.5 |
Math Not comparable
Devstral Small 2505: —, o3: 48.9 (#49)
| Benchmark | Devstral Small 2505 | o3 |
|---|---|---|
| FrontierMath (Tiers 1-3) | — | 33.3% |
| OTIS Mock AIME 2024-2025 | — | 84.4% |
| Omni-MATH | — | 71.4% |
| LMArena Math | — | 1426 |
| MATH Level 5 | — | 97.8% |
| FrontierMath (Feb 2025 set) | — | 18.7% |
| FrontierMath Tier 4 (v1) | — | 2.1% |
Knowledge Not comparable
Devstral Small 2505: —, o3: 54.5 (#47)
| Benchmark | Devstral Small 2505 | o3 |
|---|---|---|
| GPQA Diamond | — | 81.8% |
| Humanity's Last Exam | — | 20.3% |
| SimpleQA Verified | — | 49.4% |
| MMLU-Pro | — | 85.9% |
| Confabulations | — | 14.4% |
| GPQA (HELM) | — | 75.3% |
| LMArena Expert | — | 1403 |
Multimodal Not comparable
Devstral Small 2505: —, o3: 40.7 (#35)
| Benchmark | Devstral Small 2505 | o3 |
|---|---|---|
| LMArena Vision | — | 1214 |
| GeoBench | — | 74% |
| VPCT | — | 52% |
Multilingual Not comparable
Devstral Small 2505: —, o3: 51.9 (#99)
| Benchmark | Devstral Small 2505 | o3 |
|---|---|---|
| LMArena Non-English | — | 1402 |
| LMArena Chinese | — | 1438 |
| LMArena French | — | 1430 |
| LMArena German | — | 1420 |
| LMArena Japanese | — | 1404 |
| LMArena Korean | — | 1371 |
| LMArena Russian | — | 1407 |
| LMArena Spanish | — | 1395 |
Instruction Following Not comparable
Devstral Small 2505: —, o3: 73.0 (#122)
| Benchmark | Devstral Small 2505 | o3 |
|---|---|---|
| IFEval | — | 86.9% |
| LMArena Instruction Following | — | 1368 |
Long Context Not comparable
Devstral Small 2505: —, o3: 53.2 (#6)
| Benchmark | Devstral Small 2505 | o3 |
|---|---|---|
| Fiction.LiveBench | — | 88.9% |
| CL-bench | — | 17.8% |
| LMArena Longer Query | — | 1372 |
Writing & Preference Not comparable
Devstral Small 2505: —, o3: 63.7 (#62)
| Benchmark | Devstral Small 2505 | o3 |
|---|---|---|
| LMArena Text | — | 1410 |
| LMArena Creative Writing | — | 1359 |
| Short-Story Creative Writing | — | 83.9% |
| EQ-Bench Creative Writing | — | 1676 |
| WildBench | — | 86.1% |
| LMArena Multi-Turn | — | 1405 |
Frequently asked questions
Is Devstral Small 2505 better than o3?
o3 is the stronger model overall, scoring 47.4 to 33.9 on the Noometry Index. Devstral Small 2505 costs 23× less per token, which makes it the better buy when o3's lead doesn't matter for your workload.
Which is cheaper, Devstral Small 2505 or o3?
Devstral Small 2505 is cheaper. It lists at $0.10 per million input tokens and $0.30 per million output tokens; o3 lists at $2 and $8.
Is Devstral Small 2505 or o3 better for coding?
o3 scores higher on coding benchmarks: 47.3 versus 38.6 in the Noometry coding category.
Which has the bigger context window?
o3 does, with 200K tokens against 128K.
How many benchmarks do Devstral Small 2505 and o3 share?
3 benchmarks have published results for both models. Devstral Small 2505 has 4 scored results on Noometry and o3 has 63.