Noometry

Model comparison

Inkling vs Qwen3.8 27B

Qwen3.8 27B is the stronger model overall, scoring 46.0 to 44.1 on the Noometry Index.

Last verified . 28 shared benchmarks.

Inkling Thinking Machines Lab

44.1

Rank #80 Confirmed

Qwen3.8 27B Alibaba (Qwen)

46.0

Rank #68 Confirmed

Summary

  • They share 28 benchmarks with published results for both. Inkling scores higher in 2 categories and Qwen3.8 27B in 7 categories; 4 gaps are clear of the uncertainty.
  • The widest gap is in coding, where Qwen3.8 27B leads 50.5 to 34.5.
  • The biggest single-benchmark swing is ProofBench: 0% for Inkling and 16% for Qwen3.8 27B.
  • Qwen3.8 27B is cheaper at $0.99 / $1.49 per million input/output tokens, against $1.87 / $4.68 for Inkling.
  • Qwen3.8 27B accepts more context: 262K tokens versus 66K.

Side by side

InklingQwen3.8 27B
ProviderThinking Machines LabAlibaba (Qwen)
Noometry Index44.146.0
Released2026-07-152026-08-14
WeightsOpenOpen
Context window66K262K
Max output66K33K
Input $ / M tokens$1.87$0.99
Output $ / M tokens$4.68$1.49
Results tracked4131

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

Category by category

Coding Qwen3.8 27B leads

Inkling: 34.5 (#234), Qwen3.8 27B: 50.5 (#44)

BenchmarkInklingQwen3.8 27B
LMArena WebDev14131593
SciCode47%46.6%
LMArena Coding14641482
FrontierCode14%—
FrontierSWE4.1%—
WeirdML32.3%—
ALE-Bench946—

Agentic & Tool Use Qwen3.8 27B leads

Inkling: 29.6 (#85), Qwen3.8 27B: 32.9 (#57)

BenchmarkInklingQwen3.8 27B
APEX-Agents33.8%47.5%
τ²-bench Banking25%—

Reasoning Too close to call

Inkling: 40.4 (#56), Qwen3.8 27B: 41.0 (#54)

BenchmarkInklingQwen3.8 27B
ARC-AGI-236.5%42.4%
ARC-AGI-179.5%87.5%
CritPt5.4%5.4%
LMArena Hard Prompts14511460
DTBench87.5%88%
LMCA37.6%41.4%
Epoch Capabilities Index148.54149.38
SimpleBench50%—
NYT Connections (extended)—54.5%
Chess Puzzles21%—
Surface Evolver Bench—45%

Math Qwen3.8 27B leads

Inkling: 31.3 (#225), Qwen3.8 27B: 37.1 (#161)

BenchmarkInklingQwen3.8 27B
ProofBench0%16%
LMArena Math14791456
FrontierMath (Tiers 1-3)33.3%—
FrontierMath Tier 44.9%—
OTIS Mock AIME 2024-202588.9%—

Knowledge Inkling leads

Inkling: 55.1 (#49), Qwen3.8 27B: 41.6 (#109)

BenchmarkInklingQwen3.8 27B
LMArena Expert14651482
GPQA Diamond88.3%—
SimpleQA Verified40.3%—

Multimodal Not comparable

Inkling: —, Qwen3.8 27B: 41.3 (#37)

BenchmarkInklingQwen3.8 27B
LMArena Vision—1271

Multilingual Too close to call

Inkling: 54.0 (#52), Qwen3.8 27B: 53.7 (#60)

BenchmarkInklingQwen3.8 27B
LMArena Non-English14341430
LMArena Chinese14901504
LMArena French14581465
LMArena German14461438
LMArena Japanese14291384
LMArena Korean14041393
LMArena Russian14291415
LMArena Spanish14481448

Instruction Following Too close to call

Inkling: 75.1 (#71), Qwen3.8 27B: 75.8 (#53)

BenchmarkInklingQwen3.8 27B
LMArena Instruction Following14261439

Long Context Too close to call

Inkling: 43.8 (#86), Qwen3.8 27B: 44.3 (#70)

BenchmarkInklingQwen3.8 27B
LMArena Longer Query14341450

Writing & Preference Too close to call

Inkling: 65.2 (#51), Qwen3.8 27B: 65.8 (#43)

BenchmarkInklingQwen3.8 27B
LMArena Text14411441
LMArena Creative Writing13871384
EQ-Bench Creative Writing16111671
LMArena Multi-Turn14361441
EQ-Bench 41226—

Frequently asked questions

Is Inkling better than Qwen3.8 27B?

Qwen3.8 27B is the stronger model overall, scoring 46.0 to 44.1 on the Noometry Index.

Which is cheaper, Inkling or Qwen3.8 27B?

Qwen3.8 27B is cheaper. It lists at $0.99 per million input tokens and $1.49 per million output tokens; Inkling lists at $1.87 and $4.68.

Is Inkling or Qwen3.8 27B better for coding?

Qwen3.8 27B scores higher on coding benchmarks: 50.5 versus 34.5 in the Noometry coding category.

Which has the bigger context window?

Qwen3.8 27B does, with 262K tokens against 66K.

How many benchmarks do Inkling and Qwen3.8 27B share?

28 benchmarks have published results for both models. Inkling has 41 scored results on Noometry and Qwen3.8 27B has 31.

Related comparisons

Go deeper