Model benchmarks
unsloth/LFM2.5-24B-A2B-GGUF local LLM performance
As of October 2026, unsloth/LFM2.5-24B-A2B-GGUF runs at up to 61.5 tok/s for local inference (best of 2 community benchmark runs across 1 GPU).
Model size
24B
Peak speed
61.5 tok/s
Average speed
61.2 tok/s
Avg PP
556.6 tok/s
Min memory
n/a
Max context
32,768 tokens
Avg output / run
32,870 tokens
Avg runtime / run
9m 9s
Avg quality
69.6
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 | 63.2 | 69.6 | 76.0 |
| Agent Workflow | 79.1 | 81.9 | 84.8 |
| Code Generation | 13.6 | 37.3 | 61.0 |
| Role Play & Narrative | 79.8 | 80.7 | 81.6 |
| Research & Analysis | 78.2 | 78.4 | 78.6 |
Performance by hardware and tool
Every hardware/tool/quantization combination unsloth/LFM2.5-24B-A2B-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 |
|---|---|---|---|---|---|---|---|---|
| NVIDIA GeForce RTX 3070 Ti | Unsloth Studio | UD-Q4_K_XL | 61.5 tok/s | 61.2 tok/s | n/a | 32,768 tokens | 69.6 | 2 |
Benchmark runs
All 2 unsloth/LFM2.5-24B-A2B-GGUF runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is unsloth/LFM2.5-24B-A2B-GGUF good for coding?
- In our benchmarks, unsloth/LFM2.5-24B-A2B-GGUF scores 37.3/100 for coding. It runs at about 61.2 tok/s, so if you want more speed, ornith-ai/Ornith-1.5-35B-A3B-GGUF is faster (~90.8 tok/s) and still scores well for coding (83.8/100).
- Is unsloth/LFM2.5-24B-A2B-GGUF good for agentic (tool-using) tasks?
- In our benchmarks, unsloth/LFM2.5-24B-A2B-GGUF scores 81.9/100 for agentic workflows. It runs at about 61.2 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 unsloth/LFM2.5-24B-A2B-GGUF for local inference?
- Across 2 community benchmark runs, unsloth/LFM2.5-24B-A2B-GGUF reaches up to 61.5 tok/s and averages 61.2 tok/s, with the fastest results on NVIDIA GeForce RTX 3070 Ti.
- Which tools have been used to run unsloth/LFM2.5-24B-A2B-GGUF?
- Benchmarks were submitted using Unsloth Studio. Results are community-contributed and updated as new runs arrive.