Model benchmarks
Muse-Glimmer-30B-KQuant-17GB-Q4_K_M local LLM performance
As of November 2026, Muse-Glimmer-30B-KQuant-17GB-Q4_K_M runs at up to 66.0 tok/s for local inference (best of 2 community benchmark runs across 0 GPUs).
Model size
30B
Peak speed
66.0 tok/s
Average speed
65.9 tok/s
Avg PP
902.6 tok/s
Min memory
25.4 GB
Max context
131,072 tokens
Avg output / run
14,967 tokens
Avg runtime / run
3m 53s
Avg quality
83.7
Benchmark runs
2
GPUs tested
0
Quality by task
Average LLM-judged quality (0–100) with the run-to-run spread shown as a P5–P95 band, overall and for each benchmark task, across all 2 runs. The low and high columns show how much the judge’s score varies between runs, and need at least two runs to display.
| Task | P5 (low) | Avg | P95 (high) |
|---|---|---|---|
| Overall | 82.2 | 83.7 | 85.2 |
| Agent Workflow | 90.0 | 90.6 | 91.2 |
| Code Generation | 71.8 | 74.0 | 76.2 |
| Role Play & Narrative | 87.2 | 89.9 | 92.5 |
| Research & Analysis | 80.0 | 80.4 | 80.9 |
Performance by hardware and tool
Every hardware/tool/quantization combination Muse-Glimmer-30B-KQuant-17GB-Q4_K_M has been benchmarked on, ranked by peak token generation speed. Last updated November 2026.
| Hardware | Tool | Quant | Best tok/s | Avg tok/s | Memory | Context | Quality | Runs |
|---|---|---|---|---|---|---|---|---|
| CPU only | llama.cpp | Q4_K_M | 66.0 tok/s | 65.9 tok/s | 25.4 GB | 131,072 tokens | 83.7 | 2 |
Benchmark runs
All 2 Muse-Glimmer-30B-KQuant-17GB-Q4_K_M runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is Muse-Glimmer-30B-KQuant-17GB-Q4_K_M good for coding?
- In our benchmarks, Muse-Glimmer-30B-KQuant-17GB-Q4_K_M scores 74.0/100 for coding. It runs at about 65.9 tok/s, so if you want more speed, turbo is faster (~124.1 tok/s) and still scores well for coding (85.3/100).
- Is Muse-Glimmer-30B-KQuant-17GB-Q4_K_M good for agentic (tool-using) tasks?
- In our benchmarks, Muse-Glimmer-30B-KQuant-17GB-Q4_K_M scores 90.6/100 for agentic workflows. It runs at about 65.9 tok/s, so if you want more speed, Qwen3.8-27B-IQ5R-V4-DFlash2-TP2-i8KV-r21 is faster (~133.3 tok/s) and still scores well for agentic workflows (91.8/100).
- How fast is Muse-Glimmer-30B-KQuant-17GB-Q4_K_M for local inference?
- Across 2 community benchmark runs, Muse-Glimmer-30B-KQuant-17GB-Q4_K_M reaches up to 66.0 tok/s and averages 65.9 tok/s.
- How much memory does Muse-Glimmer-30B-KQuant-17GB-Q4_K_M need?
- The leanest observed configuration used about 25.4 GB of memory (quantizations tested: Q4_K_M).
- Which tools have been used to run Muse-Glimmer-30B-KQuant-17GB-Q4_K_M?
- Benchmarks were submitted using llama.cpp. Results are community-contributed and updated as new runs arrive.