Model comparison
Devstral Small 2505 vs Nemotron 3 Ultra
Nemotron 3 Ultra is the stronger model overall, scoring 42.5 to 34.3 on the Noometry Index. Devstral Small 2505 costs 7.0× less per token, which makes it the better buy when Nemotron 3 Ultra's lead doesn't matter for your workload.
Last verified . 2 shared benchmarks.
Mistral AI
NVIDIASummary
- They share 2 benchmarks with published results for both. Devstral Small 2505 scores higher in 1 category and Nemotron 3 Ultra in 1 category; one gap is clear of the uncertainty.
- The widest gap is in reasoning, where Nemotron 3 Ultra leads 31.0 to 19.7.
- The biggest single-benchmark swing is SciCode: 28.8% for Devstral Small 2505 and 40.3% for Nemotron 3 Ultra.
- Devstral Small 2505 is cheaper at $0.10 / $0.30 per million input/output tokens, against $0.60 / $2.40 for Nemotron 3 Ultra.
- Nemotron 3 Ultra accepts more context: 262K tokens versus 128K.
Side by side
| Devstral Small 2505 | Nemotron 3 Ultra | |
|---|---|---|
| Provider | Mistral AI | NVIDIA |
| Noometry Index | 34.3 | 42.5 |
| Released | 2025-05-07 | 2026-06-04 |
| Weights | Open | Open |
| Context window | 128K | 262K |
| Max output | 128K | 128K |
| Input $ / M tokens | $0.10 | $0.60 |
| Output $ / M tokens | $0.30 | $2.40 |
| Results tracked | 4 | 30 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Devstral Small 2505: 38.9 (#166), Nemotron 3 Ultra: 38.1 (#182)
| Benchmark | Devstral Small 2505 | Nemotron 3 Ultra |
|---|---|---|
| SciCode | 28.8% | 40.3% |
| FrontierCode | — | 13.6% |
| SWE-bench Verified (bash only) | 56.4% | — |
| WeirdML | — | 43.5% |
| LMArena Coding | — | 1468 |
Agentic & Tool Use Not comparable
Devstral Small 2505: —, Nemotron 3 Ultra: 23.4 (#126)
| Benchmark | Devstral Small 2505 | Nemotron 3 Ultra |
|---|---|---|
| APEX-Agents | — | 22.7% |
Reasoning Nemotron 3 Ultra leads
Devstral Small 2505: 19.7 (#252), Nemotron 3 Ultra: 31.0 (#82)
| Benchmark | Devstral Small 2505 | Nemotron 3 Ultra |
|---|---|---|
| CritPt | 0% | 3.1% |
| Kagi LLM Benchmark | 37.7% | — |
| Chess Puzzles | — | 12% |
| LMArena Hard Prompts | — | 1452 |
| Mystery Game Puzzles | — | 20% |
| DTBench | — | 90.1% |
| LMCA | — | 36.9% |
| Epoch Capabilities Index | — | 146.17 |
Math Not comparable
Devstral Small 2505: —, Nemotron 3 Ultra: 35.3 (#187)
| Benchmark | Devstral Small 2505 | Nemotron 3 Ultra |
|---|---|---|
| OTIS Mock AIME 2024-2025 | — | 86.7% |
| ProofBench | — | 2% |
| LMArena Math | — | 1457 |
Knowledge Not comparable
Devstral Small 2505: —, Nemotron 3 Ultra: 52.5 (#61)
| Benchmark | Devstral Small 2505 | Nemotron 3 Ultra |
|---|---|---|
| GPQA Diamond | — | 85.4% |
| LMArena Expert | — | 1472 |
Multilingual Not comparable
Devstral Small 2505: —, Nemotron 3 Ultra: 53.5 (#64)
| Benchmark | Devstral Small 2505 | Nemotron 3 Ultra |
|---|---|---|
| LMArena Non-English | — | 1427 |
| LMArena Chinese | — | 1497 |
| LMArena French | — | 1472 |
| LMArena German | — | 1471 |
| LMArena Korean | — | 1386 |
| LMArena Russian | — | 1417 |
| LMArena Spanish | — | 1454 |
Instruction Following Not comparable
Devstral Small 2505: —, Nemotron 3 Ultra: 74.7 (#88)
| Benchmark | Devstral Small 2505 | Nemotron 3 Ultra |
|---|---|---|
| LMArena Instruction Following | — | 1418 |
Long Context Not comparable
Devstral Small 2505: —, Nemotron 3 Ultra: 43.9 (#81)
| Benchmark | Devstral Small 2505 | Nemotron 3 Ultra |
|---|---|---|
| LMArena Longer Query | — | 1437 |
Writing & Preference Not comparable
Devstral Small 2505: —, Nemotron 3 Ultra: 66.3 (#36)
| Benchmark | Devstral Small 2505 | Nemotron 3 Ultra |
|---|---|---|
| LMArena Text | — | 1445 |
| LMArena Creative Writing | — | 1398 |
| EQ-Bench Creative Writing | — | 1692 |
| LMArena Multi-Turn | — | 1411 |
Frequently asked questions
Is Devstral Small 2505 better than Nemotron 3 Ultra?
Nemotron 3 Ultra is the stronger model overall, scoring 42.5 to 34.3 on the Noometry Index. Devstral Small 2505 costs 7.0× less per token, which makes it the better buy when Nemotron 3 Ultra's lead doesn't matter for your workload.
Which is cheaper, Devstral Small 2505 or Nemotron 3 Ultra?
Devstral Small 2505 is cheaper. It lists at $0.10 per million input tokens and $0.30 per million output tokens; Nemotron 3 Ultra lists at $0.60 and $2.40.
Is Devstral Small 2505 or Nemotron 3 Ultra better for coding?
They score almost the same on coding (38.9 vs 38.1); test both on your own repository before choosing.
Which has the bigger context window?
Nemotron 3 Ultra does, with 262K tokens against 128K.
How many benchmarks do Devstral Small 2505 and Nemotron 3 Ultra share?
2 benchmarks have published results for both models. Devstral Small 2505 has 4 scored results on Noometry and Nemotron 3 Ultra has 30.