Model benchmarks
mistralai/ministral-3-14b-reasoning local LLM performance
As of November 2026, mistralai/ministral-3-14b-reasoning runs at up to 22.5 tok/s for local inference (best of 2 community benchmark runs across 1 GPU).
Model size
14B
Peak speed
22.5 tok/s
Average speed
22.2 tok/s
Avg PP
444.6 tok/s
Min memory
14.2 GB
Max context
65,536 tokens
Avg output / run
44,964 tokens
Avg runtime / run
35m 15s
Avg quality
71.9
Benchmark runs
2
GPUs tested
1
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 | 70.6 | 71.9 | 73.2 |
| Agent Workflow | 81.8 | 85.4 | 89.0 |
| Code Generation | 58.1 | 60.8 | 63.4 |
| Role Play & Narrative | 68.9 | 74.2 | 79.4 |
| Research & Analysis | 66.4 | 67.3 | 68.2 |
Performance by hardware and tool
Every hardware/tool/quantization combination mistralai/ministral-3-14b-reasoning 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 |
|---|---|---|---|---|---|---|---|---|
| NVIDIA GeForce GTX 1080 Ti | LM Studio | Q4_K_M | 22.5 tok/s | 22.2 tok/s | 14.2 GB | 65,536 tokens | 71.9 | 2 |
Benchmark runs
All 2 mistralai/ministral-3-14b-reasoning runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is mistralai/ministral-3-14b-reasoning good for coding?
- In our benchmarks, mistralai/ministral-3-14b-reasoning scores 60.8/100 for coding. It runs at about 22.2 tok/s, so if you want more speed, Qwen3.8-27B-Turbo-Brilliance-50X-RI-GSQ-RCO-IQ3_XXS-mtp is faster (~37.5 tok/s) and still scores well for coding (83.8/100).
- Is mistralai/ministral-3-14b-reasoning good for agentic (tool-using) tasks?
- In our benchmarks, mistralai/ministral-3-14b-reasoning scores 85.4/100 for agentic workflows. It runs at about 22.2 tok/s, so if you want more speed, muse-glimmer:latest is faster (~33.3 tok/s) and still scores well for agentic workflows (91.6/100).
- How fast is mistralai/ministral-3-14b-reasoning for local inference?
- Across 2 community benchmark runs, mistralai/ministral-3-14b-reasoning reaches up to 22.5 tok/s and averages 22.2 tok/s, with the fastest results on NVIDIA GeForce GTX 1080 Ti.
- How much memory does mistralai/ministral-3-14b-reasoning need?
- The leanest observed configuration used about 14.2 GB of memory (quantizations tested: Q4_K_M).
- Which tools have been used to run mistralai/ministral-3-14b-reasoning?
- Benchmarks were submitted using LM Studio. Results are community-contributed and updated as new runs arrive.