Model benchmarks
bartowski/MiniCPM5-2B-GGUF local LLM performance
As of September 2026, bartowski/MiniCPM5-2B-GGUF runs at up to 154.5 tok/s for local inference (best of 3 community benchmark runs across 2 GPUs).
Model size
2B
Peak speed
154.5 tok/s
Average speed
102.3 tok/s
Avg PP
1558.8 tok/s
Min memory
n/a
Max context
131.072 tokens
Avg output / run
13.779 tokens
Avg runtime / run
7m 52s
Avg quality
48.3
Benchmark runs
3
GPUs tested
2
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 3 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 | 41.5 | 48.3 | 53.4 |
| Agent Workflow | 35.1 | 58.4 | 72.7 |
| Code Generation | 11.6 | 19.1 | 27.2 |
| Role Play & Narrative | 45.4 | 53.2 | 61.1 |
| Research & Analysis | 51.5 | 62.4 | 73.6 |
Performance by hardware and tool
Every hardware/tool/quantization combination bartowski/MiniCPM5-2B-GGUF has been benchmarked on, ranked by peak token generation speed. Last updated September 2026.
| Hardware | Tool | Quant | Best tok/s | Avg tok/s | Memory | Context | Quality | Runs |
|---|---|---|---|---|---|---|---|---|
| NVIDIA GeForce RTX 4070 | Unsloth Studio | Q4_K_M | 154.5 tok/s | 154.5 tok/s | 4.2 GB | 65.536 tokens | 50.7 | 1 |
| NVIDIA GeForce RTX 4070 | Unsloth Studio | Q6_K_L | 129.1 tok/s | 129.1 tok/s | 5.2 GB | 131.072 tokens | 40.4 | 1 |
| Intel(R) Arc(TM) 140V GPU (16GB) | Unsloth Studio | — | 23.2 tok/s | 23.2 tok/s | n/a | 65.536 tokens | 53.7 | 1 |
Benchmark runs
All 3 bartowski/MiniCPM5-2B-GGUF runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is bartowski/MiniCPM5-2B-GGUF good for coding?
- In our benchmarks, bartowski/MiniCPM5-2B-GGUF scores 19.1/100 for coding. It runs at about 102.3 tok/s, so if you want more speed, Qwen3.6-35B-A3B-Q4_K_XL is faster (~156.9 tok/s) and still scores well for coding (82.2/100).
- Is bartowski/MiniCPM5-2B-GGUF good for agentic (tool-using) tasks?
- In our benchmarks, bartowski/MiniCPM5-2B-GGUF scores 58.4/100 for agentic workflows. It runs at about 102.3 tok/s, so if you want more speed, Tiel-Coder-35B-A3B-Q4_K_S is faster (~162.4 tok/s) and still scores well for agentic workflows (87.8/100).
- How fast is bartowski/MiniCPM5-2B-GGUF for local inference?
- Across 3 community benchmark runs, bartowski/MiniCPM5-2B-GGUF reaches up to 154.5 tok/s and averages 102.3 tok/s, with the fastest results on NVIDIA GeForce RTX 4070.
- How much memory does bartowski/MiniCPM5-2B-GGUF need?
- The leanest observed configuration used about n/a of memory (quantizations tested: Q4_K_M, Q6_K_L).
- Which tools have been used to run bartowski/MiniCPM5-2B-GGUF?
- Benchmarks were submitted using Unsloth Studio. Results are community-contributed and updated as new runs arrive.