Model benchmarks
unsloth/Phi-4-mini-instruct-GGUF local LLM performance
As of October 2026, unsloth/Phi-4-mini-instruct-GGUF runs at up to 26.6 tok/s for local inference (best of 2 community benchmark runs across 1 GPU).
Model size
3.8B
Peak speed
26.6 tok/s
Average speed
26.4 tok/s
Avg PP
322.3 tok/s
Min memory
10.7 GB
Max context
131,072 tokens
Avg output / run
3,346 tokens
Avg runtime / run
2m 18s
Avg quality
40.9
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 | 37.2 | 40.9 | 44.6 |
| Agent Workflow | 51.5 | 61.3 | 71.0 |
| Code Generation | 2.8 | 5.5 | 8.2 |
| Role Play & Narrative | 45.8 | 49.1 | 52.4 |
| Research & Analysis | 36.8 | 47.8 | 58.8 |
Performance by hardware and tool
Every hardware/tool/quantization combination unsloth/Phi-4-mini-instruct-GGUF 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 |
|---|---|---|---|---|---|---|---|---|
| Intel(R) Arc(TM) 140V GPU (16GB) | Unsloth Studio | Q4_K_M | 26.6 tok/s | 26.4 tok/s | 10.7 GB | 131,072 tokens | 40.9 | 2 |
Benchmark runs
All 2 unsloth/Phi-4-mini-instruct-GGUF runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is unsloth/Phi-4-mini-instruct-GGUF good for coding?
- In our benchmarks, unsloth/Phi-4-mini-instruct-GGUF scores 5.5/100 for coding. It runs at about 26.4 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 unsloth/Phi-4-mini-instruct-GGUF good for agentic (tool-using) tasks?
- In our benchmarks, unsloth/Phi-4-mini-instruct-GGUF scores 61.3/100 for agentic workflows. It runs at about 26.4 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 unsloth/Phi-4-mini-instruct-GGUF for local inference?
- Across 2 community benchmark runs, unsloth/Phi-4-mini-instruct-GGUF reaches up to 26.6 tok/s and averages 26.4 tok/s, with the fastest results on Intel(R) Arc(TM) 140V GPU (16GB).
- How much memory does unsloth/Phi-4-mini-instruct-GGUF need?
- The leanest observed configuration used about 10.7 GB of memory (quantizations tested: Q4_K_M).
- Which tools have been used to run unsloth/Phi-4-mini-instruct-GGUF?
- Benchmarks were submitted using Unsloth Studio. Results are community-contributed and updated as new runs arrive.