Model comparison
Claude Haiku 4.5 vs Inkling-Small
Inkling-Small is the stronger model overall, scoring 46.5 to 39.5 on the Noometry Index.
Last verified . 27 shared benchmarks.
Summary
- They share 27 benchmarks with published results for both. Claude Haiku 4.5 scores higher in 2 categories and Inkling-Small in 7 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Inkling-Small leads 38.6 to 15.1.
- The biggest single-benchmark swing is ARC-AGI-1: 47.7% for Claude Haiku 4.5 and 84% for Inkling-Small.
- Inkling-Small is cheaper at $0.45 / $1.20 per million input/output tokens, against $1 / $5 for Claude Haiku 4.5.
- Inkling-Small accepts more context: 524K tokens versus 200K.
- Inkling-Small has downloadable open weights; the other is API-only.
Side by side
| Claude Haiku 4.5 | Inkling-Small | |
|---|---|---|
| Provider | Anthropic | Thinking Machines Lab |
| Noometry Index | 39.5 | 46.5 |
| Released | 2025-10-15 | 2026-07-15 |
| Weights | Proprietary | Open |
| Context window | 200K | 524K |
| Max output | 64K | 1.05M |
| Input $ / M tokens | $1 | $0.45 |
| Output $ / M tokens | $5 | $1.20 |
| Results tracked | 53 | 33 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Claude Haiku 4.5: 44.0 (#78), Inkling-Small: 43.6 (#85)
| Benchmark | Claude Haiku 4.5 | Inkling-Small |
|---|---|---|
| LMArena WebDev | 1330 | 1409 |
| SciCode | 43.3% | 48.7% |
| LMArena Coding | 1453 | 1451 |
| SWE-bench Verified (bash only) | 66.6% | — |
| SWE-bench Multilingual | 64.7% | — |
| WeirdML | 45.4% | — |
| ALE-Bench | 653.48 | — |
Agentic & Tool Use Not comparable
Claude Haiku 4.5: 33.6 (#52), Inkling-Small: —
| Benchmark | Claude Haiku 4.5 | Inkling-Small |
|---|---|---|
| Terminal-Bench | 35.5% | — |
| Berkeley Function Calling Leaderboard | 68.7% | — |
| DeepResearch Bench | 45.5% | — |
| BALROG | 31.2% | — |
| ExploitBench | 13.7% | — |
| Vending-Bench 2 | 458.89 | — |
Reasoning Inkling-Small leads
Claude Haiku 4.5: 15.1 (#320), Inkling-Small: 38.6 (#63)
| Benchmark | Claude Haiku 4.5 | Inkling-Small |
|---|---|---|
| ARC-AGI-2 | 4% | 40.1% |
| ARC-AGI-1 | 47.7% | 84% |
| CritPt | 0% | 8.3% |
| Chess Puzzles | 8% | 18% |
| LMArena Hard Prompts | 1420 | 1423 |
| Epoch Capabilities Index | 142.41 | 150.15 |
| NYT Connections (extended) | 14.3% | — |
| Mystery Game Puzzles | — | 6% |
| DTBench | 73.6% | — |
| LMCA | 30.9% | — |
| ForecastBench | 61.4 | — |
Math Too close to call
Claude Haiku 4.5: 44.9 (#78), Inkling-Small: 45.1 (#77)
| Benchmark | Claude Haiku 4.5 | Inkling-Small |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 66.7% | 90% |
| LMArena Math | 1396 | 1459 |
| FrontierMath (Tiers 1-3) | — | 46.3% |
| FrontierMath Tier 4 | — | 17.1% |
| ProofBench | — | 6% |
| Omni-MATH | 56.1% | — |
| MATH Level 5 | 96.4% | — |
| FrontierMath (Feb 2025 set) | 5.9% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge Inkling-Small leads
Claude Haiku 4.5: 37.7 (#153), Inkling-Small: 48.2 (#77)
| Benchmark | Claude Haiku 4.5 | Inkling-Small |
|---|---|---|
| GPQA Diamond | 71.2% | 88.5% |
| SimpleQA Verified | 13.2% | 19.1% |
| LMArena Expert | 1442 | 1442 |
| MMLU-Pro | 77.7% | — |
| Vectara Hallucination Rate | 9.8% | — |
| GPQA (HELM) | 60.5% | — |
Multimodal Inkling-Small leads
Claude Haiku 4.5: 26.8 (#118), Inkling-Small: 39.1 (#62)
| Benchmark | Claude Haiku 4.5 | Inkling-Small |
|---|---|---|
| LMArena Vision | — | 1235 |
| Blueprint-Bench 2 | 0% | — |
| LMArena Document | 1420 | — |
Multilingual Inkling-Small leads
Claude Haiku 4.5: 49.9 (#129), Inkling-Small: 51.7 (#104)
| Benchmark | Claude Haiku 4.5 | Inkling-Small |
|---|---|---|
| LMArena Non-English | 1377 | 1402 |
| LMArena Chinese | 1417 | 1465 |
| LMArena French | 1408 | 1436 |
| LMArena German | 1375 | 1405 |
| LMArena Japanese | 1339 | 1405 |
| LMArena Korean | 1347 | 1363 |
| LMArena Russian | 1381 | 1391 |
| LMArena Spanish | 1420 | 1428 |
Instruction Following Inkling-Small leads
Claude Haiku 4.5: 71.4 (#149), Inkling-Small: 73.8 (#114)
| Benchmark | Claude Haiku 4.5 | Inkling-Small |
|---|---|---|
| LMArena Instruction Following | 1414 | 1399 |
| IFEval | 80.1% | — |
Long Context Too close to call
Claude Haiku 4.5: 43.6 (#92), Inkling-Small: 42.7 (#118)
| Benchmark | Claude Haiku 4.5 | Inkling-Small |
|---|---|---|
| LMArena Longer Query | 1427 | 1401 |
Writing & Preference Inkling-Small leads
Claude Haiku 4.5: 57.9 (#123), Inkling-Small: 59.6 (#107)
| Benchmark | Claude Haiku 4.5 | Inkling-Small |
|---|---|---|
| LMArena Text | 1396 | 1414 |
| LMArena Creative Writing | 1372 | 1331 |
| LMArena Multi-Turn | 1409 | 1418 |
| EQ-Bench Creative Writing | — | 1491 |
| WildBench | 83.9% | — |
| EQ-Bench 4 | 1064 | — |
Frequently asked questions
Is Claude Haiku 4.5 better than Inkling-Small?
Inkling-Small is the stronger model overall, scoring 46.5 to 39.5 on the Noometry Index.
Which is cheaper, Claude Haiku 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 Haiku 4.5 lists at $1 and $5.
Is Claude Haiku 4.5 or Inkling-Small better for coding?
They score almost the same on coding (44.0 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 Haiku 4.5 and Inkling-Small share?
27 benchmarks have published results for both models. Claude Haiku 4.5 has 53 scored results on Noometry and Inkling-Small has 33.