Model benchmarks
mistral-small:Q5 local LLM performance
As of October 2026, mistral-small:Q5 runs at up to 39.3 tok/s for local inference (best of 2 community benchmark runs across 0 GPUs).
Model size
Unknown
Peak speed
39.3 tok/s
Average speed
39.3 tok/s
Avg PP
1787.5 tok/s
Min memory
n/a
Max context
65,536 tokens
Avg output / run
4,894 tokens
Avg runtime / run
2m 10s
Avg quality
68.9
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 | 66.1 | 68.9 | 71.7 |
| Agent Workflow | 76.7 | 82.4 | 88.2 |
| Code Generation | 55.4 | 57.3 | 59.2 |
| Role Play & Narrative | 56.2 | 68.6 | 81.0 |
| Research & Analysis | 62.1 | 67.3 | 72.4 |
Performance by hardware and tool
Every hardware/tool/quantization combination mistral-small:Q5 has been benchmarked on, ranked by peak token generation speed. Last updated October 2026.
| Hardware | Tool | Quant | Best tok/s | Avg tok/s | Memory | Context | Quality | Runs |
|---|---|---|---|---|---|---|---|---|
| AMD Radeon Graphics | llama.cpp | — | 39.3 tok/s | 39.3 tok/s | n/a | 65,536 tokens | 68.9 | 2 |
Benchmark runs
All 2 mistral-small:Q5 runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is mistral-small:Q5 good for coding?
- In our benchmarks, mistral-small:Q5 scores 57.3/100 for coding. It runs at about 39.3 tok/s, so if you want more speed, local-qwen is faster (~60.2 tok/s) and still scores well for coding (84.8/100).
- Is mistral-small:Q5 good for agentic (tool-using) tasks?
- In our benchmarks, mistral-small:Q5 scores 82.4/100 for agentic workflows. It runs at about 39.3 tok/s, so if you want more speed, granite-4.2-8b-Q4_K_L is faster (~71.6 tok/s) and still scores well for agentic workflows (92.9/100).
- How fast is mistral-small:Q5 for local inference?
- Across 2 community benchmark runs, mistral-small:Q5 reaches up to 39.3 tok/s and averages 39.3 tok/s, with the fastest results on AMD Radeon Graphics.
- Which tools have been used to run mistral-small:Q5?
- Benchmarks were submitted using llama.cpp. Results are community-contributed and updated as new runs arrive.