Model comparison
Claude Sonnet 4.5 vs Inkling-Small
Inkling-Small is the stronger model overall, scoring 46.5 to 44.0 on the Noometry Index.
Last verified . 32 shared benchmarks.
Summary
- They share 32 benchmarks with published results for both. Claude Sonnet 4.5 scores higher in 6 categories and Inkling-Small in 3 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Inkling-Small leads 44.9 to 31.9.
- The biggest single-benchmark swing is ARC-AGI-2: 13.6% for Claude Sonnet 4.5 and 40.1% for Inkling-Small.
- Inkling-Small is cheaper at $0.45 / $1.20 per million input/output tokens, against $3 / $15 for Claude Sonnet 4.5.
- Claude Sonnet 4.5 accepts more context: 1M tokens versus 524K.
- Inkling-Small has downloadable open weights; the other is API-only.
Side by side
| Claude Sonnet 4.5 | Inkling-Small | |
|---|---|---|
| Provider | Anthropic | Thinking Machines Lab |
| Noometry Index | 44.0 | 46.5 |
| Released | 2025-09-29 | 2026-07-15 |
| Weights | Proprietary | Open |
| Context window | 1M | 524K |
| Max output | 64K | 1.05M |
| Input $ / M tokens | $3 | $0.45 |
| Output $ / M tokens | $15 | $1.20 |
| Results tracked | 73 | 33 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Sonnet 4.5 leads
Claude Sonnet 4.5: 47.8 (#54), Inkling-Small: 43.3 (#80)
| Benchmark | Claude Sonnet 4.5 | Inkling-Small |
|---|---|---|
| LMArena WebDev | 1393 | 1408 |
| SciCode | 44.7% | 48.7% |
| LMArena Coding | 1489 | 1451 |
| SWE-bench Verified | 71.3% | — |
| SWE-bench Verified (bash only) | 71.4% | — |
| SWE-bench Multilingual | 67% | — |
| GSO | 14.7% | — |
| WeirdML | 47.7% | — |
| ALE-Bench | 796.15 | — |
| AlgoTune | 1.52 | — |
Agentic & Tool Use Not comparable
Claude Sonnet 4.5: 38.1 (#30), Inkling-Small: —
| Benchmark | Claude Sonnet 4.5 | Inkling-Small |
|---|---|---|
| Terminal-Bench | 46.5% | — |
| Berkeley Function Calling Leaderboard | 73.2% | — |
| GDPval | 42.5% | — |
| Remote Labor Index | 2.1% | — |
| τ²-bench Airline | 72% | — |
| τ²-bench Banking | 25.3% | — |
| τ²-bench Retail | 72.4% | — |
| τ²-bench Telecom | 84.9% | — |
| Cybench | 60% | — |
| DeepResearch Bench | 52.6% | — |
| OSWorld | 62.9% | — |
| LMArena Search | 1159 | — |
| METR Time Horizons | 67.4% | — |
| Vending-Bench 2 | 3,839 | — |
Reasoning Inkling-Small leads
Claude Sonnet 4.5: 26.9 (#113), Inkling-Small: 39.0 (#53)
| Benchmark | Claude Sonnet 4.5 | Inkling-Small |
|---|---|---|
| ARC-AGI-2 | 13.6% | 40.1% |
| ARC-AGI-1 | 63.7% | 84% |
| CritPt | 1.1% | 8.3% |
| Chess Puzzles | 12% | 18% |
| LMArena Hard Prompts | 1462 | 1424 |
| Mystery Game Puzzles | 17% | 6% |
| Epoch Capabilities Index | 146.84 | 150.13 |
| SimpleBench | 54.3% | — |
| Kagi LLM Benchmark | 57.9% | — |
| NYT Connections (extended) | 37.3% | — |
| EnigmaEval | 6% | — |
| EBR-Bench | 2.4% | — |
| DTBench | 83.2% | — |
| LMCA | 38.8% | — |
| ForecastBench | 61.9 | — |
Math Inkling-Small leads
Claude Sonnet 4.5: 31.9 (#190), Inkling-Small: 44.9 (#65)
| Benchmark | Claude Sonnet 4.5 | Inkling-Small |
|---|---|---|
| FrontierMath (Tiers 1-3) | 23.9% | 46.3% |
| FrontierMath Tier 4 | 2.4% | 17.1% |
| OTIS Mock AIME 2024-2025 | 77.8% | 90% |
| ProofBench | 19% | 6% |
| LMArena Math | 1448 | 1453 |
| Omni-MATH | 55.3% | — |
| MATH Level 5 | 97.7% | — |
| FrontierMath (Feb 2025 set) | 15.2% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |
Knowledge Too close to call
Claude Sonnet 4.5: 48.4 (#70), Inkling-Small: 48.3 (#71)
| Benchmark | Claude Sonnet 4.5 | Inkling-Small |
|---|---|---|
| GPQA Diamond | 82.3% | 88.5% |
| SimpleQA Verified | 30.7% | 19.1% |
| LMArena Expert | 1483 | 1442 |
| Humanity's Last Exam | 13.7% | — |
| MMLU-Pro | 86.9% | — |
| Vectara Hallucination Rate | 12% | — |
| GPQA (HELM) | 68.6% | — |
Multimodal Inkling-Small leads
Claude Sonnet 4.5: 34.1 (#84), Inkling-Small: 38.4 (#59)
| Benchmark | Claude Sonnet 4.5 | Inkling-Small |
|---|---|---|
| LMArena Vision | — | 1235 |
| VPCT | 39.8% | — |
| LMArena Document | 1450 | — |
Multilingual Claude Sonnet 4.5 leads
Claude Sonnet 4.5: 53.6 (#65), Inkling-Small: 51.7 (#102)
| Benchmark | Claude Sonnet 4.5 | Inkling-Small |
|---|---|---|
| LMArena Non-English | 1425 | 1400 |
| LMArena Chinese | 1459 | 1458 |
| LMArena French | 1459 | 1438 |
| LMArena German | 1427 | 1412 |
| LMArena Japanese | 1390 | 1399 |
| LMArena Korean | 1403 | 1355 |
| LMArena Russian | 1436 | 1392 |
| LMArena Spanish | 1457 | 1426 |
Instruction Following Claude Sonnet 4.5 leads
Claude Sonnet 4.5: 75.2 (#74), Inkling-Small: 73.9 (#112)
| Benchmark | Claude Sonnet 4.5 | Inkling-Small |
|---|---|---|
| LMArena Instruction Following | 1459 | 1397 |
| IFEval | 85% | — |
Long Context Claude Sonnet 4.5 leads
Claude Sonnet 4.5: 45.2 (#45), Inkling-Small: 42.7 (#113)
| Benchmark | Claude Sonnet 4.5 | Inkling-Small |
|---|---|---|
| LMArena Longer Query | 1476 | 1401 |
Writing & Preference Claude Sonnet 4.5 leads
Claude Sonnet 4.5: 66.7 (#33), Inkling-Small: 59.8 (#102)
| Benchmark | Claude Sonnet 4.5 | Inkling-Small |
|---|---|---|
| LMArena Text | 1439 | 1413 |
| LMArena Creative Writing | 1442 | 1330 |
| EQ-Bench Creative Writing | 1678 | 1491 |
| LMArena Multi-Turn | 1465 | 1420 |
| WildBench | 85.4% | — |
Frequently asked questions
Is Claude Sonnet 4.5 better than Inkling-Small?
Inkling-Small is the stronger model overall, scoring 46.5 to 44.0 on the Noometry Index.
Which is cheaper, Claude Sonnet 4.5 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 4.5 lists at $3 and $15.
Is Claude Sonnet 4.5 or Inkling-Small better for coding?
Claude Sonnet 4.5 scores higher on coding benchmarks: 47.8 versus 43.3 in the Noometry coding category.
Which has the bigger context window?
Claude Sonnet 4.5 does, with 1M tokens against 524K.
How many benchmarks do Claude Sonnet 4.5 and Inkling-Small share?
32 benchmarks have published results for both models. Claude Sonnet 4.5 has 73 scored results on Noometry and Inkling-Small has 33.