Model comparison
Devstral Small 2505 vs Inkling
Inkling is the stronger model overall, scoring 44.1 to 33.9 on the Noometry Index. Devstral Small 2505 costs 17× less per token, which makes it the better buy when Inkling's lead doesn't matter for your workload.
Last verified . 2 shared benchmarks.
Mistral AISummary
- They share 2 benchmarks with published results for both. Devstral Small 2505 scores higher in 1 category and Inkling in 1 category; 2 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Inkling leads 40.6 to 19.1.
- The biggest single-benchmark swing is SciCode: 28.8% for Devstral Small 2505 and 47% for Inkling.
- Devstral Small 2505 is cheaper at $0.10 / $0.30 per million input/output tokens, against $1.87 / $4.68 for Inkling.
- Devstral Small 2505 accepts more context: 128K tokens versus 66K.
Side by side
| Devstral Small 2505 | Inkling | |
|---|---|---|
| Provider | Mistral AI | Thinking Machines Lab |
| Noometry Index | 33.9 | 44.1 |
| Released | 2025-05-07 | 2026-07-15 |
| Weights | Open | Open |
| Context window | 128K | 66K |
| Max output | 128K | 66K |
| Input $ / M tokens | $0.10 | $1.87 |
| Output $ / M tokens | $0.30 | $4.68 |
| Results tracked | 4 | 41 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Devstral Small 2505 leads
Devstral Small 2505: 38.6 (#160), Inkling: 34.1 (#229)
| Benchmark | Devstral Small 2505 | Inkling |
|---|---|---|
| SciCode | 28.8% | 47% |
| FrontierCode | — | 14% |
| SWE-bench Verified (bash only) | 56.4% | — |
| LMArena WebDev | — | 1412 |
| FrontierSWE | — | 4.1% |
| WeirdML | — | 32.3% |
| LMArena Coding | — | 1463 |
| ALE-Bench | — | 946 |
Agentic & Tool Use Not comparable
Devstral Small 2505: —, Inkling: 29.5 (#76)
| Benchmark | Devstral Small 2505 | Inkling |
|---|---|---|
| APEX-Agents | — | 33.8% |
| τ²-bench Banking | — | 25% |
Reasoning Inkling leads
Devstral Small 2505: 19.1 (#243), Inkling: 40.6 (#48)
| Benchmark | Devstral Small 2505 | Inkling |
|---|---|---|
| CritPt | 0% | 5.4% |
| ARC-AGI-2 | — | 36.5% |
| SimpleBench | — | 50% |
| Kagi LLM Benchmark | 37.7% | — |
| ARC-AGI-1 | — | 79.5% |
| Chess Puzzles | — | 21% |
| LMArena Hard Prompts | — | 1452 |
| DTBench | — | 87.5% |
| LMCA | — | 37.6% |
| Epoch Capabilities Index | — | 148.6 |
Math Not comparable
Devstral Small 2505: —, Inkling: 31.2 (#198)
| Benchmark | Devstral Small 2505 | Inkling |
|---|---|---|
| FrontierMath (Tiers 1-3) | — | 33.3% |
| FrontierMath Tier 4 | — | 4.9% |
| OTIS Mock AIME 2024-2025 | — | 88.9% |
| ProofBench | — | 0% |
| LMArena Math | — | 1482 |
Knowledge Not comparable
Devstral Small 2505: —, Inkling: 55.2 (#44)
| Benchmark | Devstral Small 2505 | Inkling |
|---|---|---|
| GPQA Diamond | — | 88.3% |
| SimpleQA Verified | — | 40.3% |
| LMArena Expert | — | 1468 |
Multilingual Not comparable
Devstral Small 2505: —, Inkling: 54.2 (#49)
| Benchmark | Devstral Small 2505 | Inkling |
|---|---|---|
| LMArena Non-English | — | 1434 |
| LMArena Chinese | — | 1495 |
| LMArena French | — | 1453 |
| LMArena German | — | 1456 |
| LMArena Japanese | — | 1428 |
| LMArena Korean | — | 1401 |
| LMArena Russian | — | 1430 |
| LMArena Spanish | — | 1442 |
Instruction Following Not comparable
Devstral Small 2505: —, Inkling: 75.3 (#65)
| Benchmark | Devstral Small 2505 | Inkling |
|---|---|---|
| LMArena Instruction Following | — | 1427 |
Long Context Not comparable
Devstral Small 2505: —, Inkling: 43.9 (#79)
| Benchmark | Devstral Small 2505 | Inkling |
|---|---|---|
| LMArena Longer Query | — | 1436 |
Writing & Preference Not comparable
Devstral Small 2505: —, Inkling: 65.5 (#48)
| Benchmark | Devstral Small 2505 | Inkling |
|---|---|---|
| LMArena Text | — | 1443 |
| LMArena Creative Writing | — | 1390 |
| EQ-Bench Creative Writing | — | 1611 |
| EQ-Bench 4 | — | 1226 |
| LMArena Multi-Turn | — | 1436 |
Frequently asked questions
Is Devstral Small 2505 better than Inkling?
Inkling is the stronger model overall, scoring 44.1 to 33.9 on the Noometry Index. Devstral Small 2505 costs 17× less per token, which makes it the better buy when Inkling's lead doesn't matter for your workload.
Which is cheaper, Devstral Small 2505 or Inkling?
Devstral Small 2505 is cheaper. It lists at $0.10 per million input tokens and $0.30 per million output tokens; Inkling lists at $1.87 and $4.68.
Is Devstral Small 2505 or Inkling better for coding?
Devstral Small 2505 scores higher on coding benchmarks: 38.6 versus 34.1 in the Noometry coding category.
Which has the bigger context window?
Devstral Small 2505 does, with 128K tokens against 66K.
How many benchmarks do Devstral Small 2505 and Inkling share?
2 benchmarks have published results for both models. Devstral Small 2505 has 4 scored results on Noometry and Inkling has 41.