# Claude 3.5 Haiku vs Inkling

> Inkling is the stronger model overall, scoring 44.1 to 28.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-3-5-haiku-vs-inkling
- Last updated: 2026-10-10
- Shared benchmarks: 25

## Summary

- They share 25 benchmarks with published results for both. Claude 3.5 Haiku scores higher in 0 categories and Inkling in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Inkling leads 55.2 to 18.7.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 4.3% for Claude 3.5 Haiku and 88.9% for Inkling.
- Inkling has downloadable open weights; the other is API-only.

## Snapshot

| | Claude 3.5 Haiku | Inkling |
|---|---|---|
| Provider | Anthropic | Thinking Machines Lab |
| Noometry Index | 28.9 | 44.1 |
| Rank | 309 | 75 |
| Context | — | 66K |
| Input $/M | — | $1.87 |
| Output $/M | — | $4.68 |
| Weights | Proprietary | Open |

## Coding

- Claude 3.5 Haiku: 32.6 (#258)
- Inkling: 34.1 (#229)

| Benchmark | Claude 3.5 Haiku | Inkling |
|---|---|---|
| SciCode | 27.4% | 47% |
| WeirdML | 30.7% | 32.3% |
| LMArena Coding | 1287 | 1463 |
| FrontierCode | — | 14% |
| Aider Polyglot | 28% | — |
| LMArena WebDev | — | 1412 |
| FrontierSWE | — | 4.1% |
| BigCodeBench Instruct | 46.1% | — |
| LiveBench Coding | 51.4% | — |
| BigCodeBench Complete | 59% | — |
| CadEval | 32% | — |
| ALE-Bench | — | 946 |

## Agentic & Tool Use

- Claude 3.5 Haiku: 27.7 (#90)
- Inkling: 29.5 (#76)

| Benchmark | Claude 3.5 Haiku | Inkling |
|---|---|---|
| APEX-Agents | — | 33.8% |
| τ²-bench Banking | — | 25% |
| BALROG | 19.3% | — |

## Reasoning

- Claude 3.5 Haiku: 17.1 (#286)
- Inkling: 40.6 (#48)

| Benchmark | Claude 3.5 Haiku | Inkling |
|---|---|---|
| CritPt | 0% | 5.4% |
| LMArena Hard Prompts | 1251 | 1452 |
| DTBench | 56.7% | 87.5% |
| Epoch Capabilities Index | 127.15 | 148.6 |
| ARC-AGI-2 | — | 36.5% |
| SimpleBench | — | 50% |
| ARC-AGI-1 | — | 79.5% |
| Chess Puzzles | — | 21% |
| LiveBench Reasoning | 28.1% | — |
| LiveBench Data Analysis | 48.5% | — |
| LMCA | — | 37.6% |
| LiveBench | 43.5% | — |

## Math

- Claude 3.5 Haiku: 13.7 (#294)
- Inkling: 31.2 (#198)

| Benchmark | Claude 3.5 Haiku | Inkling |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 4.3% | 88.9% |
| LMArena Math | 1244 | 1482 |
| FrontierMath (Tiers 1-3) | — | 33.3% |
| FrontierMath Tier 4 | — | 4.9% |
| ProofBench | — | 0% |
| Omni-MATH | 22.4% | — |
| LiveBench Math | 35.5% | — |
| MATH Level 5 | 46.4% | — |
| FrontierMath (Feb 2025 set) | 0.3% | — |

## Knowledge

- Claude 3.5 Haiku: 18.7 (#276)
- Inkling: 55.2 (#44)

| Benchmark | Claude 3.5 Haiku | Inkling |
|---|---|---|
| GPQA Diamond | 38.1% | 88.3% |
| LMArena Expert | 1208 | 1468 |
| SimpleQA Verified | — | 40.3% |
| MMLU-Pro | 60.5% | — |
| Confabulations | 36.7% | — |
| GPQA (HELM) | 36.3% | — |
| MMLU | 74.3% | — |

## Multimodal

- Claude 3.5 Haiku: 26.3 (#113)
- Inkling: —

| Benchmark | Claude 3.5 Haiku | Inkling |
|---|---|---|
| LMArena Vision | 1092 | — |
| GeoBench | 34% | — |

## Multilingual

- Claude 3.5 Haiku: 40.2 (#213)
- Inkling: 54.2 (#49)

| Benchmark | Claude 3.5 Haiku | Inkling |
|---|---|---|
| LMArena Non-English | 1238 | 1434 |
| LMArena Chinese | 1229 | 1495 |
| LMArena French | 1262 | 1453 |
| LMArena German | 1237 | 1456 |
| LMArena Japanese | 1175 | 1428 |
| LMArena Korean | 1172 | 1401 |
| LMArena Russian | 1253 | 1430 |
| LMArena Spanish | 1261 | 1442 |

## Instruction Following

- Claude 3.5 Haiku: 63.1 (#229)
- Inkling: 75.3 (#65)

| Benchmark | Claude 3.5 Haiku | Inkling |
|---|---|---|
| LMArena Instruction Following | 1241 | 1427 |
| LiveBench Instruction Following | 61.9% | — |
| IFEval | 79.2% | — |

## Long Context

- Claude 3.5 Haiku: 38.3 (#195)
- Inkling: 43.9 (#79)

| Benchmark | Claude 3.5 Haiku | Inkling |
|---|---|---|
| LMArena Longer Query | 1261 | 1436 |

## Writing & Preference

- Claude 3.5 Haiku: 42.9 (#228)
- Inkling: 65.5 (#48)

| Benchmark | Claude 3.5 Haiku | Inkling |
|---|---|---|
| LMArena Text | 1255 | 1443 |
| LMArena Creative Writing | 1233 | 1390 |
| EQ-Bench Creative Writing | 1146 | 1611 |
| LMArena Multi-Turn | 1265 | 1436 |
| Short-Story Creative Writing | 73.5% | — |
| WildBench | 76% | — |
| EQ-Bench 4 | — | 1226 |
| LiveBench Language | 35.4% | — |

## FAQ

### Is Claude 3.5 Haiku better than Inkling?

Inkling is the stronger model overall, scoring 44.1 to 28.9 on the Noometry Index.

### Is Claude 3.5 Haiku or Inkling better for coding?

Inkling scores higher on coding benchmarks: 34.1 versus 32.6 in the Noometry coding category.

### How many benchmarks do Claude 3.5 Haiku and Inkling share?

25 benchmarks have published results for both models. Claude 3.5 Haiku has 49 scored results on Noometry and Inkling has 41.