Noometry

Model comparison

Claude 3 Haiku vs Inkling

Inkling is the stronger model overall, scoring 44.1 to 25.9 on the Noometry Index.

Last verified . 24 shared benchmarks.

Claude 3 Haiku Anthropic

25.9

Rank #340 Confirmed

Inkling Thinking Machines Lab

44.1

Rank #80 Confirmed

Summary

  • They share 24 benchmarks with published results for both. Claude 3 Haiku scores higher in 0 categories and Inkling in 8 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in knowledge, where Inkling leads 55.1 to 17.3.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 1.8% for Claude 3 Haiku and 88.9% for Inkling.
  • Inkling has downloadable open weights; the other is API-only.

Side by side

Claude 3 HaikuInkling
ProviderAnthropicThinking Machines Lab
Noometry Index25.944.1
Released2024-03-072026-07-15
WeightsProprietaryOpen
Context window—66K
Max output—66K
Input $ / M tokens—$1.87
Output $ / M tokens—$4.68
Results tracked3741

Sponsored placements are available on pages like this one. Advertise on Noometry

Category by category

Coding Inkling leads

Claude 3 Haiku: 26.4 (#325), Inkling: 34.5 (#234)

BenchmarkClaude 3 HaikuInkling
WeirdML9.8%32.3%
LMArena Coding11991464
FrontierCode—14%
LMArena WebDev—1413
FrontierSWE—4.1%
SciCode—47%
BigCodeBench Instruct39.4%—
BigCodeBench Complete50.1%—
CadEval12%—
ALE-Bench—946
HumanEval+68.9%—
MBPP+68.8%—

Agentic & Tool Use Not comparable

Claude 3 Haiku: —, Inkling: 29.6 (#85)

BenchmarkClaude 3 HaikuInkling
APEX-Agents—33.8%
τ²-bench Banking—25%

Reasoning Inkling leads

Claude 3 Haiku: 16.3 (#307), Inkling: 40.4 (#56)

BenchmarkClaude 3 HaikuInkling
LMArena Hard Prompts11741451
DTBench50.1%87.5%
LMCA8.8%37.6%
Epoch Capabilities Index118.35148.54
ARC-AGI-2—36.5%
SimpleBench—50%
Kagi LLM Benchmark34.2%—
ARC-AGI-1—79.5%
CritPt—5.4%
Chess Puzzles—21%
ForecastBench53.2—
WinoGrande74.2%—

Math Inkling leads

Claude 3 Haiku: 9.8 (#319), Inkling: 31.3 (#225)

BenchmarkClaude 3 HaikuInkling
OTIS Mock AIME 2024-20251.8%88.9%
LMArena Math11881479
FrontierMath (Tiers 1-3)—33.3%
FrontierMath Tier 4—4.9%
ProofBench—0%
MATH Level 514.9%—

Knowledge Inkling leads

Claude 3 Haiku: 17.3 (#285), Inkling: 55.1 (#49)

BenchmarkClaude 3 HaikuInkling
GPQA Diamond36.3%88.3%
LMArena Expert11481465
SimpleQA Verified—40.3%
Confabulations34.2%—
MMLU73.8%—

Multimodal Not comparable

Claude 3 Haiku: 23.6 (#128), Inkling: —

BenchmarkClaude 3 HaikuInkling
LMArena Vision950—
ScienceQA72%—

Multilingual Inkling leads

Claude 3 Haiku: 36.0 (#243), Inkling: 54.0 (#52)

BenchmarkClaude 3 HaikuInkling
LMArena Non-English11781434
LMArena Chinese11551490
LMArena French11951458
LMArena German11741446
LMArena Japanese11021429
LMArena Korean11091404
LMArena Russian12041429
LMArena Spanish11661448

Instruction Following Inkling leads

Claude 3 Haiku: 61.3 (#247), Inkling: 75.1 (#71)

BenchmarkClaude 3 HaikuInkling
LMArena Instruction Following11731426

Long Context Inkling leads

Claude 3 Haiku: 36.1 (#237), Inkling: 43.8 (#86)

BenchmarkClaude 3 HaikuInkling
LMArena Longer Query11901434

Writing & Preference Inkling leads

Claude 3 Haiku: 29.7 (#291), Inkling: 65.2 (#51)

BenchmarkClaude 3 HaikuInkling
LMArena Text11951441
LMArena Creative Writing11571387
EQ-Bench Creative Writing7171611
LMArena Multi-Turn11901436
EQ-Bench 4—1226

Frequently asked questions

Is Claude 3 Haiku better than Inkling?

Inkling is the stronger model overall, scoring 44.1 to 25.9 on the Noometry Index.

Is Claude 3 Haiku or Inkling better for coding?

Inkling scores higher on coding benchmarks: 34.5 versus 26.4 in the Noometry coding category.

How many benchmarks do Claude 3 Haiku and Inkling share?

24 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and Inkling has 41.

Related comparisons

Go deeper