Model comparison
Claude Sonnet 4 vs Inkling-Small
Inkling-Small is the stronger model overall, scoring 46.5 to 40.8 on the Noometry Index.
Last verified . 26 shared benchmarks.
Summary
- They share 26 benchmarks with published results for both. Claude Sonnet 4 scores higher in 0 categories and Inkling-Small in 9 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Inkling-Small leads 38.6 to 22.9.
- The biggest single-benchmark swing is ARC-AGI-1: 40% for Claude Sonnet 4 and 84% for Inkling-Small.
- Inkling-Small is cheaper at $0.45 / $1.20 per million input/output tokens, against $3 / $15 for Claude Sonnet 4.
- Inkling-Small accepts more context: 524K tokens versus 200K.
- Inkling-Small has downloadable open weights; the other is API-only.
Side by side
| Claude Sonnet 4 | Inkling-Small | |
|---|---|---|
| Provider | Anthropic | Thinking Machines Lab |
| Noometry Index | 40.8 | 46.5 |
| Released | 2025-05-22 | 2026-07-15 |
| Weights | Proprietary | Open |
| Context window | 200K | 524K |
| Max output | 64K | 1.05M |
| Input $ / M tokens | $3 | $0.45 |
| Output $ / M tokens | $15 | $1.20 |
| Results tracked | 58 | 33 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Claude Sonnet 4: 43.5 (#88), Inkling-Small: 43.6 (#85)
| Benchmark | Claude Sonnet 4 | Inkling-Small |
|---|---|---|
| SciCode | 40% | 48.7% |
| LMArena Coding | 1414 | 1451 |
| SWE-bench Verified (bash only) | 64.9% | — |
| Aider Polyglot | 61.3% | — |
| LMArena WebDev | — | 1409 |
| GSO | 4.9% | — |
| WeirdML | 46.1% | — |
| ALE-Bench | 655.35 | — |
Agentic & Tool Use Not comparable
Claude Sonnet 4: 38.5 (#31), Inkling-Small: —
| Benchmark | Claude Sonnet 4 | Inkling-Small |
|---|---|---|
| TheAgentCompany | 33.1% | — |
| Cybench | 35% | — |
| DeepResearch Bench | 46.6% | — |
| OSWorld | 43.9% | — |
| METR Time Horizons | 62% | — |
Reasoning Inkling-Small leads
Claude Sonnet 4: 22.9 (#187), Inkling-Small: 38.6 (#63)
| Benchmark | Claude Sonnet 4 | Inkling-Small |
|---|---|---|
| ARC-AGI-2 | 5.9% | 40.1% |
| ARC-AGI-1 | 40% | 84% |
| CritPt | 0.3% | 8.3% |
| LMArena Hard Prompts | 1372 | 1423 |
| Epoch Capabilities Index | 141.69 | 150.15 |
| SimpleBench | 45.5% | — |
| Kagi LLM Benchmark | 73% | — |
| Chess Puzzles | — | 18% |
| EnigmaEval | 3.1% | — |
| Mystery Game Puzzles | — | 6% |
| DTBench | 77.1% | — |
| LMCA | 29% | — |
| ForecastBench | 60.2 | — |
Math Inkling-Small leads
Claude Sonnet 4: 43.3 (#80), Inkling-Small: 45.1 (#77)
| Benchmark | Claude Sonnet 4 | Inkling-Small |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 71.1% | 90% |
| LMArena Math | 1375 | 1459 |
| FrontierMath (Tiers 1-3) | — | 46.3% |
| FrontierMath Tier 4 | — | 17.1% |
| ProofBench | — | 6% |
| Omni-MATH | 60.2% | — |
| MATH Level 5 | 84.4% | — |
| FrontierMath (Feb 2025 set) | 4.1% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge Inkling-Small leads
Claude Sonnet 4: 41.8 (#108), Inkling-Small: 48.2 (#77)
| Benchmark | Claude Sonnet 4 | Inkling-Small |
|---|---|---|
| GPQA Diamond | 79.2% | 88.5% |
| LMArena Expert | 1372 | 1442 |
| Humanity's Last Exam | 7.8% | — |
| SimpleQA Verified | — | 19.1% |
| MMLU-Pro | 84.3% | — |
| Confabulations | 13.2% | — |
| Vectara Hallucination Rate | 10.3% | — |
| GPQA (HELM) | 70.6% | — |
Multimodal Inkling-Small leads
Claude Sonnet 4: 26.2 (#121), Inkling-Small: 39.1 (#62)
| Benchmark | Claude Sonnet 4 | Inkling-Small |
|---|---|---|
| LMArena Vision | 1191 | 1235 |
| GeoBench | 37% | — |
| VPCT | 34% | — |
| MindCube | 44.8% | — |
Multilingual Inkling-Small leads
Claude Sonnet 4: 46.7 (#156), Inkling-Small: 51.7 (#104)
| Benchmark | Claude Sonnet 4 | Inkling-Small |
|---|---|---|
| LMArena Non-English | 1333 | 1402 |
| LMArena Chinese | 1350 | 1465 |
| LMArena French | 1363 | 1436 |
| LMArena German | 1331 | 1405 |
| LMArena Japanese | 1302 | 1405 |
| LMArena Korean | 1291 | 1363 |
| LMArena Russian | 1355 | 1391 |
| LMArena Spanish | 1357 | 1428 |
Instruction Following Inkling-Small leads
Claude Sonnet 4: 71.7 (#145), Inkling-Small: 73.8 (#114)
| Benchmark | Claude Sonnet 4 | Inkling-Small |
|---|---|---|
| LMArena Instruction Following | 1376 | 1399 |
| IFEval | 84% | — |
Long Context Inkling-Small leads
Claude Sonnet 4: 33.7 (#259), Inkling-Small: 42.7 (#118)
| Benchmark | Claude Sonnet 4 | Inkling-Small |
|---|---|---|
| LMArena Longer Query | 1398 | 1401 |
| Fiction.LiveBench | 46.9% | — |
Writing & Preference Inkling-Small leads
Claude Sonnet 4: 57.1 (#132), Inkling-Small: 59.6 (#107)
| Benchmark | Claude Sonnet 4 | Inkling-Small |
|---|---|---|
| LMArena Text | 1351 | 1414 |
| LMArena Creative Writing | 1345 | 1331 |
| EQ-Bench Creative Writing | 1483 | 1491 |
| LMArena Multi-Turn | 1376 | 1418 |
| Short-Story Creative Writing | 81.4% | — |
| WildBench | 83.8% | — |
Frequently asked questions
Is Claude Sonnet 4 better than Inkling-Small?
Inkling-Small is the stronger model overall, scoring 46.5 to 40.8 on the Noometry Index.
Which is cheaper, Claude Sonnet 4 or Inkling-Small?
Inkling-Small is cheaper. It lists at $0.45 per million input tokens and $1.20 per million output tokens; Claude Sonnet 4 lists at $3 and $15.
Is Claude Sonnet 4 or Inkling-Small better for coding?
They score almost the same on coding (43.5 vs 43.6); test both on your own repository before choosing.
Which has the bigger context window?
Inkling-Small does, with 524K tokens against 200K.
How many benchmarks do Claude Sonnet 4 and Inkling-Small share?
26 benchmarks have published results for both models. Claude Sonnet 4 has 58 scored results on Noometry and Inkling-Small has 33.