# Devstral Small 2505 vs GPT-5 Nano

> Devstral Small 2505 and GPT-5 Nano score almost the same on the Noometry Index (34.3 vs 33.5), so choose on price, context window or the category you care about most.

- Canonical page: https://noometry.com/compare/devstral-small-vs-gpt-5-nano
- Last updated: 2026-10-11
- Shared benchmarks: 2

## Summary

- They share 2 benchmarks with published results for both. Devstral Small 2505 scores higher in 2 categories and GPT-5 Nano in 0 categories; 2 gaps are clear of the uncertainty.
- The widest gap is in coding, where Devstral Small 2505 leads 38.9 to 33.6.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 37.7% for Devstral Small 2505 and 62.2% for GPT-5 Nano.
- GPT-5 Nano is cheaper at $0.05 / $0.40 per million input/output tokens, against $0.10 / $0.30 for Devstral Small 2505.
- GPT-5 Nano accepts more context: 400K tokens versus 128K.
- Devstral Small 2505 has downloadable open weights; the other is API-only.

## Snapshot

| | Devstral Small 2505 | GPT-5 Nano |
|---|---|---|
| Provider | Mistral AI | OpenAI |
| Noometry Index | 34.3 | 33.5 |
| Rank | 233 | 241 |
| Context | 128K | 400K |
| Input $/M | $0.10 | $0.05 |
| Output $/M | $0.30 | $0.40 |
| Weights | Open | Proprietary |

## Coding

- Devstral Small 2505: 38.9 (#166)
- GPT-5 Nano: 33.6 (#254)

| Benchmark | Devstral Small 2505 | GPT-5 Nano |
|---|---|---|
| SWE-bench Verified (bash only) | 56.4% | 34.8% |
| SciCode | 28.8% | — |
| WeirdML | — | 38.1% |
| LMArena Coding | — | 1351 |
| ALE-Bench | — | 718.67 |

## Agentic & Tool Use

- Devstral Small 2505: —
- GPT-5 Nano: 25.8 (#106)

| Benchmark | Devstral Small 2505 | GPT-5 Nano |
|---|---|---|
| Terminal-Bench | — | 21.8% |
| Berkeley Function Calling Leaderboard | — | 51.5% |

## Reasoning

- Devstral Small 2505: 19.7 (#252)
- GPT-5 Nano: 16.3 (#306)

| Benchmark | Devstral Small 2505 | GPT-5 Nano |
|---|---|---|
| Kagi LLM Benchmark | 37.7% | 62.2% |
| ARC-AGI-2 | — | 2.6% |
| ARC-AGI-1 | — | 20.7% |
| CritPt | 0% | — |
| Chess Puzzles | — | 27% |
| LMArena Hard Prompts | — | 1328 |
| Mystery Game Puzzles | — | 9% |
| DTBench | — | 62.7% |
| LMCA | — | 7.9% |
| Epoch Capabilities Index | — | 139.38 |
| ForecastBench | — | 59.1 |

## Math

- Devstral Small 2505: —
- GPT-5 Nano: 29.4 (#241)

| Benchmark | Devstral Small 2505 | GPT-5 Nano |
|---|---|---|
| FrontierMath (Tiers 1-3) | — | 20% |
| FrontierMath Tier 4 | — | 2.4% |
| OTIS Mock AIME 2024-2025 | — | 81.1% |
| ProofBench | — | 12% |
| Omni-MATH | — | 54.6% |
| LMArena Math | — | 1317 |
| MATH Level 5 | — | 95.2% |
| FrontierMath (Feb 2025 set) | — | 8.3% |
| FrontierMath Tier 4 (v1) | — | 2.1% |

## Knowledge

- Devstral Small 2505: —
- GPT-5 Nano: 35.9 (#178)

| Benchmark | Devstral Small 2505 | GPT-5 Nano |
|---|---|---|
| GPQA Diamond | — | 69.4% |
| SimpleQA Verified | — | 11.7% |
| MMLU-Pro | — | 77.8% |
| Vectara Hallucination Rate | — | 10.5% |
| GPQA (HELM) | — | 67.9% |
| LMArena Expert | — | 1321 |

## Multimodal

- Devstral Small 2505: —
- GPT-5 Nano: 31.3 (#108)

| Benchmark | Devstral Small 2505 | GPT-5 Nano |
|---|---|---|
| LMArena Vision | — | 1159 |
| VPCT | — | 37.2% |

## Multilingual

- Devstral Small 2505: —
- GPT-5 Nano: 45.3 (#172)

| Benchmark | Devstral Small 2505 | GPT-5 Nano |
|---|---|---|
| LMArena Non-English | — | 1313 |
| LMArena Chinese | — | 1356 |
| LMArena German | — | 1327 |
| LMArena Japanese | — | 1226 |
| LMArena Korean | — | 1269 |
| LMArena Russian | — | 1296 |
| LMArena Spanish | — | 1360 |

## Instruction Following

- Devstral Small 2505: —
- GPT-5 Nano: 75.0 (#79)

| Benchmark | Devstral Small 2505 | GPT-5 Nano |
|---|---|---|
| IFEval | — | 93.2% |
| LMArena Instruction Following | — | 1306 |

## Long Context

- Devstral Small 2505: —
- GPT-5 Nano: 31.3 (#281)

| Benchmark | Devstral Small 2505 | GPT-5 Nano |
|---|---|---|
| Fiction.LiveBench | — | 44.4% |
| LMArena Longer Query | — | 1312 |

## Writing & Preference

- Devstral Small 2505: —
- GPT-5 Nano: 39.1 (#249)

| Benchmark | Devstral Small 2505 | GPT-5 Nano |
|---|---|---|
| LMArena Text | — | 1320 |
| LMArena Creative Writing | — | 1249 |
| EQ-Bench Creative Writing | — | 705 |
| WildBench | — | 80.6% |
| LMArena Multi-Turn | — | 1311 |

## FAQ

### Is Devstral Small 2505 better than GPT-5 Nano?

Devstral Small 2505 and GPT-5 Nano score almost the same on the Noometry Index (34.3 vs 33.5), so choose on price, context window or the category you care about most.

### Which is cheaper, Devstral Small 2505 or GPT-5 Nano?

GPT-5 Nano is cheaper. It lists at $0.05 per million input tokens and $0.40 per million output tokens; Devstral Small 2505 lists at $0.10 and $0.30.

### Is Devstral Small 2505 or GPT-5 Nano better for coding?

Devstral Small 2505 scores higher on coding benchmarks: 38.9 versus 33.6 in the Noometry coding category.

### Which has the bigger context window?

GPT-5 Nano does, with 400K tokens against 128K.

### How many benchmarks do Devstral Small 2505 and GPT-5 Nano share?

2 benchmarks have published results for both models. Devstral Small 2505 has 4 scored results on Noometry and GPT-5 Nano has 49.