Model comparison
Muse Glimmer vs Qwen2.5-VL 72B Instruct
Muse Glimmer is the stronger model overall, scoring 41.7 to 29.9 on the Noometry Index.
Last verified . 0 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- The widest gap is in reasoning, where Muse Glimmer leads 25.7 to 20.7.
Side by side
| Muse Glimmer | Qwen2.5-VL 72B Instruct | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 41.7 | 29.9 |
| Released | 2026-08-10 | 2024-09 |
| Weights | Open | Open |
| Context window | — | 131K |
| Max output | — | 8K |
| Input $ / M tokens | — | $2.80 |
| Output $ / M tokens | — | $8.40 |
| Results tracked | 15 | 6 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Muse Glimmer: 40.4 (#142), Qwen2.5-VL 72B Instruct: —
| Benchmark | Muse Glimmer | Qwen2.5-VL 72B Instruct |
|---|---|---|
| LMArena WebDev | 1355 | — |
| SciCode | 44.9% | — |
| LMArena Coding | 1416 | — |
Agentic & Tool Use Not comparable
Muse Glimmer: —, Qwen2.5-VL 72B Instruct: 18.6 (#144)
| Benchmark | Muse Glimmer | Qwen2.5-VL 72B Instruct |
|---|---|---|
| OSWorld | — | 5% |
Reasoning Muse Glimmer leads
Muse Glimmer: 25.7 (#144), Qwen2.5-VL 72B Instruct: 20.7 (#233)
| Benchmark | Muse Glimmer | Qwen2.5-VL 72B Instruct |
|---|---|---|
| Kagi LLM Benchmark | — | 36% |
| CritPt | 2.6% | — |
| LMArena Hard Prompts | 1396 | — |
Math Not comparable
Muse Glimmer: 38.8 (#125), Qwen2.5-VL 72B Instruct: —
| Benchmark | Muse Glimmer | Qwen2.5-VL 72B Instruct |
|---|---|---|
| LMArena Math | 1417 | — |
Knowledge Not comparable
Muse Glimmer: 38.5 (#143), Qwen2.5-VL 72B Instruct: —
| Benchmark | Muse Glimmer | Qwen2.5-VL 72B Instruct |
|---|---|---|
| LMArena Expert | 1386 | — |
Multimodal Not comparable
Muse Glimmer: —, Qwen2.5-VL 72B Instruct: 33.5 (#97)
| Benchmark | Muse Glimmer | Qwen2.5-VL 72B Instruct |
|---|---|---|
| LMArena Vision | — | 1107 |
| Video-MME | — | 73.5% |
| GeoBench | — | 62% |
| SpatialViz-Bench | — | 33.3% |
Multilingual Not comparable
Muse Glimmer: 50.5 (#122), Qwen2.5-VL 72B Instruct: —
| Benchmark | Muse Glimmer | Qwen2.5-VL 72B Instruct |
|---|---|---|
| LMArena Non-English | 1384 | — |
| LMArena Chinese | 1412 | — |
| LMArena Russian | 1390 | — |
Instruction Following Not comparable
Muse Glimmer: 72.6 (#135), Qwen2.5-VL 72B Instruct: —
| Benchmark | Muse Glimmer | Qwen2.5-VL 72B Instruct |
|---|---|---|
| LMArena Instruction Following | 1375 | — |
Long Context Not comparable
Muse Glimmer: 42.1 (#129), Qwen2.5-VL 72B Instruct: —
| Benchmark | Muse Glimmer | Qwen2.5-VL 72B Instruct |
|---|---|---|
| LMArena Longer Query | 1382 | — |
Writing & Preference Not comparable
Muse Glimmer: 57.5 (#126), Qwen2.5-VL 72B Instruct: —
| Benchmark | Muse Glimmer | Qwen2.5-VL 72B Instruct |
|---|---|---|
| LMArena Text | 1389 | — |
| LMArena Creative Writing | 1339 | — |
| LMArena Multi-Turn | 1399 | — |
Frequently asked questions
Is Muse Glimmer better than Qwen2.5-VL 72B Instruct?
Muse Glimmer is the stronger model overall, scoring 41.7 to 29.9 on the Noometry Index.
How many benchmarks do Muse Glimmer and Qwen2.5-VL 72B Instruct share?
0 benchmarks have published results for both models. Muse Glimmer has 15 scored results on Noometry and Qwen2.5-VL 72B Instruct has 6.