Model comparison
Devstral Small 2505 vs Step 3.5 Flash
Step 3.5 Flash is the stronger model overall, scoring 42.3 to 34.3 on the Noometry Index.
Last verified . 0 shared benchmarks.
Summary
- The widest gap is in coding, where Step 3.5 Flash leads 42.4 to 38.9.
- Both cost about the same: $0.10 input and $0.30 output per million tokens.
- Step 3.5 Flash accepts more context: 256K tokens versus 128K.
Side by side
| Devstral Small 2505 | Step 3.5 Flash | |
|---|---|---|
| Provider | Mistral AI | StepFun |
| Noometry Index | 34.3 | 42.3 |
| Released | 2025-05-07 | 2026-01-29 |
| Weights | Open | Open |
| Context window | 128K | 256K |
| Max output | 128K | 256K |
| Input $ / M tokens | $0.10 | $0.10 |
| Output $ / M tokens | $0.30 | $0.30 |
| Results tracked | 4 | 19 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 3.5 Flash leads
Devstral Small 2505: 38.9 (#166), Step 3.5 Flash: 42.4 (#105)
| Benchmark | Devstral Small 2505 | Step 3.5 Flash |
|---|---|---|
| SWE-bench Verified (bash only) | 56.4% | — |
| SciCode | 28.8% | — |
| LMArena Coding | — | 1436 |
Reasoning Step 3.5 Flash leads
Devstral Small 2505: 19.7 (#252), Step 3.5 Flash: 22.2 (#202)
| Benchmark | Devstral Small 2505 | Step 3.5 Flash |
|---|---|---|
| Kagi LLM Benchmark | 37.7% | — |
| NYT Connections (extended) | — | 28.4% |
| CritPt | 0% | — |
| LMArena Hard Prompts | — | 1411 |
Math Not comparable
Devstral Small 2505: —, Step 3.5 Flash: 42.6 (#84)
| Benchmark | Devstral Small 2505 | Step 3.5 Flash |
|---|---|---|
| MathArena Final-Answer Competitions | — | 66.8% |
| LMArena Math | — | 1408 |
Knowledge Not comparable
Devstral Small 2505: —, Step 3.5 Flash: 39.6 (#132)
| Benchmark | Devstral Small 2505 | Step 3.5 Flash |
|---|---|---|
| LMArena Expert | — | 1421 |
Multilingual Not comparable
Devstral Small 2505: —, Step 3.5 Flash: 50.5 (#119)
| Benchmark | Devstral Small 2505 | Step 3.5 Flash |
|---|---|---|
| LMArena Non-English | — | 1385 |
| LMArena Chinese | — | 1447 |
| LMArena French | — | 1421 |
| LMArena German | — | 1405 |
| LMArena Japanese | — | 1354 |
| LMArena Korean | — | 1352 |
| LMArena Russian | — | 1385 |
| LMArena Spanish | — | 1419 |
Instruction Following Not comparable
Devstral Small 2505: —, Step 3.5 Flash: 73.1 (#124)
| Benchmark | Devstral Small 2505 | Step 3.5 Flash |
|---|---|---|
| LMArena Instruction Following | — | 1385 |
Long Context Not comparable
Devstral Small 2505: —, Step 3.5 Flash: 42.8 (#117)
| Benchmark | Devstral Small 2505 | Step 3.5 Flash |
|---|---|---|
| LMArena Longer Query | — | 1402 |
Writing & Preference Not comparable
Devstral Small 2505: —, Step 3.5 Flash: 58.8 (#113)
| Benchmark | Devstral Small 2505 | Step 3.5 Flash |
|---|---|---|
| LMArena Text | — | 1403 |
| LMArena Creative Writing | — | 1357 |
| LMArena Multi-Turn | — | 1405 |
Frequently asked questions
Is Devstral Small 2505 better than Step 3.5 Flash?
Step 3.5 Flash is the stronger model overall, scoring 42.3 to 34.3 on the Noometry Index.
Which is cheaper, Devstral Small 2505 or Step 3.5 Flash?
Step 3.5 Flash is cheaper. It lists at $0.10 per million input tokens and $0.30 per million output tokens; Devstral Small 2505 lists at $0.10 and $0.30.
Is Devstral Small 2505 or Step 3.5 Flash better for coding?
Step 3.5 Flash scores higher on coding benchmarks: 42.4 versus 38.9 in the Noometry coding category.
Which has the bigger context window?
Step 3.5 Flash does, with 256K tokens against 128K.
How many benchmarks do Devstral Small 2505 and Step 3.5 Flash share?
0 benchmarks have published results for both models. Devstral Small 2505 has 4 scored results on Noometry and Step 3.5 Flash has 19.