Model comparison
Claude 3.7 Sonnet vs Inkling
Inkling is the stronger model overall, scoring 44.1 to 39.5 on the Noometry Index.
Last verified . 24 shared benchmarks.
Summary
- They share 24 benchmarks with published results for both. Claude 3.7 Sonnet scores higher in 4 categories and Inkling in 5 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Inkling leads 40.4 to 18.6.
- The biggest single-benchmark swing is ARC-AGI-1: 28.6% for Claude 3.7 Sonnet and 79.5% for Inkling.
- Inkling has downloadable open weights; the other is API-only.
Side by side
| Claude 3.7 Sonnet | Inkling | |
|---|---|---|
| Provider | Anthropic | Thinking Machines Lab |
| Noometry Index | 39.5 | 44.1 |
| Released | 2025-02-24 | 2026-07-15 |
| Weights | Proprietary | Open |
| Context window | — | 66K |
| Max output | — | 66K |
| Input $ / M tokens | — | $1.87 |
| Output $ / M tokens | — | $4.68 |
| Results tracked | 58 | 41 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 40.6 (#136), Inkling: 34.5 (#234)
| Benchmark | Claude 3.7 Sonnet | Inkling |
|---|---|---|
| LMArena Coding | 1361 | 1464 |
| SWE-bench Verified | 61% | — |
| FrontierCode | — | 14% |
| SWE-bench Verified (bash only) | 52.8% | — |
| Aider Polyglot | 64.9% | — |
| LMArena WebDev | — | 1413 |
| FrontierSWE | — | 4.1% |
| SciCode | — | 47% |
| GSO | 3.8% | — |
| WeirdML | — | 32.3% |
| LiveBench Coding | 74.5% | — |
| CadEval | 54% | — |
| ALE-Bench | — | 946 |
Agentic & Tool Use Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 34.1 (#50), Inkling: 29.6 (#85)
| Benchmark | Claude 3.7 Sonnet | Inkling |
|---|---|---|
| APEX-Agents | — | 33.8% |
| TheAgentCompany | 30.9% | — |
| τ²-bench Banking | — | 25% |
| Cybench | 20% | — |
| DeepResearch Bench | 43.6% | — |
| OSWorld | 35.8% | — |
| METR Time Horizons | 60% | — |
Reasoning Inkling leads
Claude 3.7 Sonnet: 18.6 (#277), Inkling: 40.4 (#56)
| Benchmark | Claude 3.7 Sonnet | Inkling |
|---|---|---|
| ARC-AGI-2 | 0.9% | 36.5% |
| SimpleBench | 46.4% | 50% |
| ARC-AGI-1 | 28.6% | 79.5% |
| LMArena Hard Prompts | 1333 | 1451 |
| Epoch Capabilities Index | 141.16 | 148.54 |
| CritPt | — | 5.4% |
| Chess Puzzles | — | 21% |
| EnigmaEval | 4.2% | — |
| LiveBench Reasoning | 87.8% | — |
| DTBench | — | 87.5% |
| LiveBench Data Analysis | 74% | — |
| LMCA | — | 37.6% |
| ForecastBench | 61.8 | — |
| LiveBench | 76.1% | — |
Math Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 37.5 (#153), Inkling: 31.3 (#225)
| Benchmark | Claude 3.7 Sonnet | Inkling |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 57.8% | 88.9% |
| LMArena Math | 1337 | 1479 |
| FrontierMath (Tiers 1-3) | — | 33.3% |
| FrontierMath Tier 4 | — | 4.9% |
| ProofBench | — | 0% |
| Omni-MATH | 33% | — |
| LiveBench Math | 79% | — |
| MATH Level 5 | 91.2% | — |
| FrontierMath (Feb 2025 set) | 4.1% | — |
Knowledge Inkling leads
Claude 3.7 Sonnet: 39.8 (#130), Inkling: 55.1 (#49)
| Benchmark | Claude 3.7 Sonnet | Inkling |
|---|---|---|
| GPQA Diamond | 79.7% | 88.3% |
| LMArena Expert | 1321 | 1465 |
| Humanity's Last Exam | 8% | — |
| SimpleQA Verified | — | 40.3% |
| MMLU-Pro | 78.4% | — |
| Confabulations | 14.7% | — |
| GPQA (HELM) | 60.8% | — |
Multimodal Not comparable
Claude 3.7 Sonnet: 33.7 (#95), Inkling: —
| Benchmark | Claude 3.7 Sonnet | Inkling |
|---|---|---|
| LMArena Vision | 1169 | — |
| GeoBench | 68% | — |
| VPCT | 39% | — |
| SpatialViz-Bench | 33.9% | — |
Multilingual Inkling leads
Claude 3.7 Sonnet: 44.1 (#179), Inkling: 54.0 (#52)
| Benchmark | Claude 3.7 Sonnet | Inkling |
|---|---|---|
| LMArena Non-English | 1296 | 1434 |
| LMArena Chinese | 1299 | 1490 |
| LMArena French | 1303 | 1458 |
| LMArena German | 1301 | 1446 |
| LMArena Japanese | 1267 | 1429 |
| LMArena Korean | 1249 | 1404 |
| LMArena Russian | 1311 | 1429 |
| LMArena Spanish | 1298 | 1448 |
Instruction Following Inkling leads
Claude 3.7 Sonnet: 72.9 (#125), Inkling: 75.1 (#71)
| Benchmark | Claude 3.7 Sonnet | Inkling |
|---|---|---|
| LMArena Instruction Following | 1352 | 1426 |
| LiveBench Instruction Following | 81.3% | — |
| IFEval | 83.4% | — |
Long Context Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 50.3 (#10), Inkling: 43.8 (#86)
| Benchmark | Claude 3.7 Sonnet | Inkling |
|---|---|---|
| LMArena Longer Query | 1373 | 1434 |
| Fiction.LiveBench | 83.3% | — |
Writing & Preference Inkling leads
Claude 3.7 Sonnet: 54.4 (#150), Inkling: 65.2 (#51)
| Benchmark | Claude 3.7 Sonnet | Inkling |
|---|---|---|
| LMArena Text | 1314 | 1441 |
| LMArena Creative Writing | 1332 | 1387 |
| EQ-Bench Creative Writing | 1412 | 1611 |
| LMArena Multi-Turn | 1339 | 1436 |
| Short-Story Creative Writing | 81.1% | — |
| WildBench | 81.4% | — |
| EQ-Bench 4 | — | 1226 |
| LiveBench Language | 59.9% | — |
Frequently asked questions
Is Claude 3.7 Sonnet better than Inkling?
Inkling is the stronger model overall, scoring 44.1 to 39.5 on the Noometry Index.
Is Claude 3.7 Sonnet or Inkling better for coding?
Claude 3.7 Sonnet scores higher on coding benchmarks: 40.6 versus 34.5 in the Noometry coding category.
How many benchmarks do Claude 3.7 Sonnet and Inkling share?
24 benchmarks have published results for both models. Claude 3.7 Sonnet has 58 scored results on Noometry and Inkling has 41.