Noometry

Model comparison

Inkling-Small vs Qwen2.5-VL 72B Instruct

Inkling-Small is the stronger model overall, scoring 46.5 to 29.4 on the Noometry Index.

Last verified . 1 shared benchmarks.

Inkling-Small Thinking Machines Lab

46.5

Rank #57 Confirmed

Qwen2.5-VL 72B Instruct Alibaba (Qwen)

29.4

Rank #298 Confirmed

Summary

  • They share 1 benchmark with published results for both. Inkling-Small scores higher in 2 categories and Qwen2.5-VL 72B Instruct in 0 categories; 2 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Inkling-Small leads 39.0 to 20.0.
  • Inkling-Small is cheaper at $0.45 / $1.20 per million input/output tokens, against $2.80 / $8.40 for Qwen2.5-VL 72B Instruct.
  • Inkling-Small accepts more context: 524K tokens versus 131K.

Side by side

Inkling-SmallQwen2.5-VL 72B Instruct
ProviderThinking Machines LabAlibaba (Qwen)
Noometry Index46.529.4
Released2026-07-152024-09
WeightsOpenOpen
Context window524K131K
Max output1.05M8K
Input $ / M tokens$0.45$2.80
Output $ / M tokens$1.20$8.40
Results tracked336

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

Category by category

Coding Not comparable

Inkling-Small: 43.3 (#80), Qwen2.5-VL 72B Instruct: —

BenchmarkInkling-SmallQwen2.5-VL 72B Instruct
LMArena WebDev1408—
SciCode48.7%—
LMArena Coding1451—

Agentic & Tool Use Not comparable

Inkling-Small: —, Qwen2.5-VL 72B Instruct: 18.4 (#136)

BenchmarkInkling-SmallQwen2.5-VL 72B Instruct
OSWorld—5%

Reasoning Inkling-Small leads

Inkling-Small: 39.0 (#53), Qwen2.5-VL 72B Instruct: 20.0 (#228)

BenchmarkInkling-SmallQwen2.5-VL 72B Instruct
ARC-AGI-240.1%—
Kagi LLM Benchmark—36%
ARC-AGI-184%—
CritPt8.3%—
Chess Puzzles18%—
LMArena Hard Prompts1424—
Mystery Game Puzzles6%—
Epoch Capabilities Index150.13—

Math Not comparable

Inkling-Small: 44.9 (#65), Qwen2.5-VL 72B Instruct: —

BenchmarkInkling-SmallQwen2.5-VL 72B Instruct
FrontierMath (Tiers 1-3)46.3%—
FrontierMath Tier 417.1%—
OTIS Mock AIME 2024-202590%—
ProofBench6%—
LMArena Math1453—

Knowledge Not comparable

Inkling-Small: 48.3 (#71), Qwen2.5-VL 72B Instruct: —

BenchmarkInkling-SmallQwen2.5-VL 72B Instruct
GPQA Diamond88.5%—
SimpleQA Verified19.1%—
LMArena Expert1442—

Multimodal Inkling-Small leads

Inkling-Small: 38.4 (#59), Qwen2.5-VL 72B Instruct: 32.9 (#92)

BenchmarkInkling-SmallQwen2.5-VL 72B Instruct
LMArena Vision12351107
Video-MME—73.5%
GeoBench—62%
SpatialViz-Bench—33.3%

Multilingual Not comparable

Inkling-Small: 51.7 (#102), Qwen2.5-VL 72B Instruct: —

BenchmarkInkling-SmallQwen2.5-VL 72B Instruct
LMArena Non-English1400—
LMArena Chinese1458—
LMArena French1438—
LMArena German1412—
LMArena Japanese1399—
LMArena Korean1355—
LMArena Russian1392—
LMArena Spanish1426—

Instruction Following Not comparable

Inkling-Small: 73.9 (#112), Qwen2.5-VL 72B Instruct: —

BenchmarkInkling-SmallQwen2.5-VL 72B Instruct
LMArena Instruction Following1397—

Long Context Not comparable

Inkling-Small: 42.7 (#113), Qwen2.5-VL 72B Instruct: —

BenchmarkInkling-SmallQwen2.5-VL 72B Instruct
LMArena Longer Query1401—

Writing & Preference Not comparable

Inkling-Small: 59.8 (#102), Qwen2.5-VL 72B Instruct: —

BenchmarkInkling-SmallQwen2.5-VL 72B Instruct
LMArena Text1413—
LMArena Creative Writing1330—
EQ-Bench Creative Writing1491—
LMArena Multi-Turn1420—

Frequently asked questions

Is Inkling-Small better than Qwen2.5-VL 72B Instruct?

Inkling-Small is the stronger model overall, scoring 46.5 to 29.4 on the Noometry Index.

Which is cheaper, Inkling-Small or Qwen2.5-VL 72B Instruct?

Inkling-Small is cheaper. It lists at $0.45 per million input tokens and $1.20 per million output tokens; Qwen2.5-VL 72B Instruct lists at $2.80 and $8.40.

Which has the bigger context window?

Inkling-Small does, with 524K tokens against 131K.

How many benchmarks do Inkling-Small and Qwen2.5-VL 72B Instruct share?

1 benchmark has published results for both models. Inkling-Small has 33 scored results on Noometry and Qwen2.5-VL 72B Instruct has 6.

Related comparisons

Go deeper