Compare AI models

Specifications
Parameters
30B0.8B/2B/4B/9B
Context window
128k256k
API pricing
Cheapest input
$0.30DeepInfra—
Cheapest output
$1.20DeepInfra—
$120ModelRun · bf16—
Benchmarks
GPQA Diamond
83.5%81.7%
IFBench
77%64.5%
OSWorld-Verified
65.9%41.8%
CharXiv Reasoning
78.8%73%
MMMU-Pro
74%70.1%
Overview
CompanyMetaQwen
Release dateAug 10 2026Mar 2 2026
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Muse Glimmer leads Qwen3.5 Small on 5 of the 5 benchmarks they both report. Qwen3.5 Small shipped 161 days before Muse Glimmer, so benchmark comparisons should account for the intervening progress.

Muse Glimmer has 30B parameters, while Qwen3.5 Small has 0.8B/2B/4B/9B. Context windows are 128k (Muse Glimmer) vs 256k (Qwen3.5 Small).

On GPQA Diamond, Muse Glimmer leads at 83.5% vs Qwen3.5 Small at 81.7%. On IFBench, Muse Glimmer leads at 77% vs Qwen3.5 Small at 64.5%. On OSWorld-Verified, Muse Glimmer leads at 65.9% vs Qwen3.5 Small at 41.8%. On CharXiv Reasoning, Muse Glimmer leads at 78.8% vs Qwen3.5 Small at 73%. On MMMU-Pro, Muse Glimmer leads at 74% vs Qwen3.5 Small at 70.1%.