Noometry

Model comparison

Inkling vs Qwen2.5-VL 72B Instruct

Inkling is the stronger model overall, scoring 44.1 to 29.4 on the Noometry Index.

Last verified . 0 shared benchmarks.

Inkling Thinking Machines Lab

44.1

Rank #75 Confirmed

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

29.4

Rank #298 Confirmed

Summary

  • The widest gap is in reasoning, where Inkling leads 40.6 to 20.0.
  • Inkling is cheaper at $1.87 / $4.68 per million input/output tokens, against $2.80 / $8.40 for Qwen2.5-VL 72B Instruct.
  • Qwen2.5-VL 72B Instruct accepts more context: 131K tokens versus 66K.

Side by side

InklingQwen2.5-VL 72B Instruct
ProviderThinking Machines LabAlibaba (Qwen)
Noometry Index44.129.4
Released2026-07-152024-09
WeightsOpenOpen
Context window66K131K
Max output66K8K
Input $ / M tokens$1.87$2.80
Output $ / M tokens$4.68$8.40
Results tracked416

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

Category by category

Coding Not comparable

Inkling: 34.1 (#229), Qwen2.5-VL 72B Instruct: —

BenchmarkInklingQwen2.5-VL 72B Instruct
FrontierCode14%—
LMArena WebDev1412—
FrontierSWE4.1%—
SciCode47%—
WeirdML32.3%—
LMArena Coding1463—
ALE-Bench946—

Agentic & Tool Use Inkling leads

Inkling: 29.5 (#76), Qwen2.5-VL 72B Instruct: 18.4 (#136)

BenchmarkInklingQwen2.5-VL 72B Instruct
APEX-Agents33.8%—
τ²-bench Banking25%—
OSWorld—5%

Reasoning Inkling leads

Inkling: 40.6 (#48), Qwen2.5-VL 72B Instruct: 20.0 (#228)

BenchmarkInklingQwen2.5-VL 72B Instruct
ARC-AGI-236.5%—
SimpleBench50%—
Kagi LLM Benchmark—36%
ARC-AGI-179.5%—
CritPt5.4%—
Chess Puzzles21%—
LMArena Hard Prompts1452—
DTBench87.5%—
LMCA37.6%—
Epoch Capabilities Index148.6—

Math Not comparable

Inkling: 31.2 (#198), Qwen2.5-VL 72B Instruct: —

BenchmarkInklingQwen2.5-VL 72B Instruct
FrontierMath (Tiers 1-3)33.3%—
FrontierMath Tier 44.9%—
OTIS Mock AIME 2024-202588.9%—
ProofBench0%—
LMArena Math1482—

Knowledge Not comparable

Inkling: 55.2 (#44), Qwen2.5-VL 72B Instruct: —

BenchmarkInklingQwen2.5-VL 72B Instruct
GPQA Diamond88.3%—
SimpleQA Verified40.3%—
LMArena Expert1468—

Multimodal Not comparable

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

BenchmarkInklingQwen2.5-VL 72B Instruct
LMArena Vision—1107
Video-MME—73.5%
GeoBench—62%
SpatialViz-Bench—33.3%

Multilingual Not comparable

Inkling: 54.2 (#49), Qwen2.5-VL 72B Instruct: —

BenchmarkInklingQwen2.5-VL 72B Instruct
LMArena Non-English1434—
LMArena Chinese1495—
LMArena French1453—
LMArena German1456—
LMArena Japanese1428—
LMArena Korean1401—
LMArena Russian1430—
LMArena Spanish1442—

Instruction Following Not comparable

Inkling: 75.3 (#65), Qwen2.5-VL 72B Instruct: —

BenchmarkInklingQwen2.5-VL 72B Instruct
LMArena Instruction Following1427—

Long Context Not comparable

Inkling: 43.9 (#79), Qwen2.5-VL 72B Instruct: —

BenchmarkInklingQwen2.5-VL 72B Instruct
LMArena Longer Query1436—

Writing & Preference Not comparable

Inkling: 65.5 (#48), Qwen2.5-VL 72B Instruct: —

BenchmarkInklingQwen2.5-VL 72B Instruct
LMArena Text1443—
LMArena Creative Writing1390—
EQ-Bench Creative Writing1611—
EQ-Bench 41226—
LMArena Multi-Turn1436—

Frequently asked questions

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

Inkling is the stronger model overall, scoring 44.1 to 29.4 on the Noometry Index.

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

Inkling is cheaper. It lists at $1.87 per million input tokens and $4.68 per million output tokens; Qwen2.5-VL 72B Instruct lists at $2.80 and $8.40.

Which has the bigger context window?

Qwen2.5-VL 72B Instruct does, with 131K tokens against 66K.

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

0 benchmarks have published results for both models. Inkling has 41 scored results on Noometry and Qwen2.5-VL 72B Instruct has 6.

Related comparisons

Go deeper