Model comparison
Devstral Small 2505 vs Gemini 2.5 Flash
Gemini 2.5 Flash is the stronger model overall, scoring 39.3 to 34.3 on the Noometry Index. Devstral Small 2505 costs 5.7× less per token, which makes it the better buy when Gemini 2.5 Flash's lead doesn't matter for your workload.
Last verified . 3 shared benchmarks.
Mistral AI
GoogleSummary
- They share 3 benchmarks with published results for both. Devstral Small 2505 scores higher in 2 categories and Gemini 2.5 Flash in 0 categories; 2 gaps are clear of the uncertainty.
- The widest gap is in coding, where Devstral Small 2505 leads 38.9 to 35.8.
- The biggest single-benchmark swing is SWE-bench Verified (bash only): 56.4% for Devstral Small 2505 and 28.7% for Gemini 2.5 Flash.
- Devstral Small 2505 is cheaper at $0.10 / $0.30 per million input/output tokens, against $0.30 / $2.50 for Gemini 2.5 Flash.
- Gemini 2.5 Flash 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 | Gemini 2.5 Flash | |
|---|---|---|
| Provider | Mistral AI | |
| Noometry Index | 34.3 | 39.3 |
| Released | 2025-05-07 | 2025-04-17 |
| Weights | Open | Proprietary |
| Context window | 128K | 1.05M |
| Max output | 128K | 66K |
| Input $ / M tokens | $0.10 | $0.30 |
| Output $ / M tokens | $0.30 | $2.50 |
| Results tracked | 4 | 54 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Devstral Small 2505 leads
Devstral Small 2505: 38.9 (#166), Gemini 2.5 Flash: 35.8 (#220)
| Benchmark | Devstral Small 2505 | Gemini 2.5 Flash |
|---|---|---|
| SWE-bench Verified (bash only) | 56.4% | 28.7% |
| Aider Polyglot | — | 55.1% |
| SciCode | 28.8% | — |
| WeirdML | — | 41.9% |
| LMArena Coding | — | 1424 |
| ALE-Bench | — | 661.88 |
Agentic & Tool Use Not comparable
Devstral Small 2505: —, Gemini 2.5 Flash: 30.8 (#74)
| Benchmark | Devstral Small 2505 | Gemini 2.5 Flash |
|---|---|---|
| Terminal-Bench | — | 17.1% |
| Berkeley Function Calling Leaderboard | — | 56.2% |
| TheAgentCompany | — | 41.1% |
| BALROG | — | 33.5% |
| Vending-Bench 2 | — | 548.84 |
Reasoning Devstral Small 2505 leads
Devstral Small 2505: 19.7 (#252), Gemini 2.5 Flash: 18.1 (#286)
| Benchmark | Devstral Small 2505 | Gemini 2.5 Flash |
|---|---|---|
| Kagi LLM Benchmark | 37.7% | 56.8% |
| CritPt | 0% | 1.1% |
| ARC-AGI-2 | — | 2.5% |
| SimpleBench | — | 41.2% |
| ARC-AGI-1 | — | 33.3% |
| EnigmaEval | — | 2.7% |
| LMArena Hard Prompts | — | 1422 |
| DTBench | — | 76.5% |
| LMCA | — | 27.5% |
| Epoch Capabilities Index | — | 143.03 |
| ForecastBench | — | 60.6 |
Math Not comparable
Devstral Small 2505: —, Gemini 2.5 Flash: 39.9 (#98)
| Benchmark | Devstral Small 2505 | Gemini 2.5 Flash |
|---|---|---|
| OTIS Mock AIME 2024-2025 | — | 73.1% |
| Omni-MATH | — | 38.5% |
| LMArena Math | — | 1415 |
| FrontierMath (Feb 2025 set) | — | 4.8% |
| FrontierMath Tier 4 (v1) | — | 4.2% |
Knowledge Not comparable
Devstral Small 2505: —, Gemini 2.5 Flash: 36.4 (#168)
| Benchmark | Devstral Small 2505 | Gemini 2.5 Flash |
|---|---|---|
| Humanity's Last Exam | — | 12.1% |
| MMLU-Pro | — | 63.9% |
| Confabulations | — | 16.8% |
| Vectara Hallucination Rate | — | 7.8% |
| GPQA (HELM) | — | 39% |
| LMArena Expert | — | 1426 |
Multimodal Not comparable
Devstral Small 2505: —, Gemini 2.5 Flash: 41.8 (#32)
| Benchmark | Devstral Small 2505 | Gemini 2.5 Flash |
|---|---|---|
| LMArena Vision | — | 1253 |
| GeoBench | — | 76% |
| VPCT | — | 46.2% |
| SpatialViz-Bench | — | 36.9% |
Multilingual Not comparable
Devstral Small 2505: —, Gemini 2.5 Flash: 52.3 (#88)
| Benchmark | Devstral Small 2505 | Gemini 2.5 Flash |
|---|---|---|
| LMArena Non-English | — | 1409 |
| LMArena Chinese | — | 1450 |
| LMArena French | — | 1433 |
| LMArena German | — | 1418 |
| LMArena Japanese | — | 1405 |
| LMArena Korean | — | 1385 |
| LMArena Russian | — | 1415 |
| LMArena Spanish | — | 1421 |
Instruction Following Not comparable
Devstral Small 2505: —, Gemini 2.5 Flash: 75.7 (#54)
| Benchmark | Devstral Small 2505 | Gemini 2.5 Flash |
|---|---|---|
| IFEval | — | 89.8% |
| LMArena Instruction Following | — | 1405 |
Long Context Not comparable
Devstral Small 2505: —, Gemini 2.5 Flash: 47.5 (#17)
| Benchmark | Devstral Small 2505 | Gemini 2.5 Flash |
|---|---|---|
| Fiction.LiveBench | — | 77.8% |
| LMArena Longer Query | — | 1419 |
Writing & Preference Not comparable
Devstral Small 2505: —, Gemini 2.5 Flash: 53.8 (#157)
| Benchmark | Devstral Small 2505 | Gemini 2.5 Flash |
|---|---|---|
| LMArena Text | — | 1417 |
| LMArena Creative Writing | — | 1400 |
| Short-Story Creative Writing | — | 76.5% |
| EQ-Bench Creative Writing | — | 1137 |
| WildBench | — | 81.7% |
| LMArena Multi-Turn | — | 1408 |
Frequently asked questions
Is Devstral Small 2505 better than Gemini 2.5 Flash?
Gemini 2.5 Flash is the stronger model overall, scoring 39.3 to 34.3 on the Noometry Index. Devstral Small 2505 costs 5.7× less per token, which makes it the better buy when Gemini 2.5 Flash's lead doesn't matter for your workload.
Which is cheaper, Devstral Small 2505 or Gemini 2.5 Flash?
Devstral Small 2505 is cheaper. It lists at $0.10 per million input tokens and $0.30 per million output tokens; Gemini 2.5 Flash lists at $0.30 and $2.50.
Is Devstral Small 2505 or Gemini 2.5 Flash better for coding?
Devstral Small 2505 scores higher on coding benchmarks: 38.9 versus 35.8 in the Noometry coding category.
Which has the bigger context window?
Gemini 2.5 Flash does, with 1.05M tokens against 128K.
How many benchmarks do Devstral Small 2505 and Gemini 2.5 Flash share?
3 benchmarks have published results for both models. Devstral Small 2505 has 4 scored results on Noometry and Gemini 2.5 Flash has 54.