Model benchmarks
liquidai/lfm2.5-2.6b local LLM performance
As of August 2026, liquidai/lfm2.5-2.6b runs at up to 165.9 tok/s for local inference (best of 2 community benchmark runs across 1 GPU).
Model size
2.6B
Peak speed
165.9 tok/s
Average speed
132.7 tok/s
Min memory
n/a
Max context
32,768 tokens
Avg runtime / run
3m 31s
Avg quality
56.8
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 | 53.9 | 56.8 | 59.8 |
| Agent Workflow | 81.3 | 81.6 | 81.9 |
| Code Generation | 16.7 | 17.3 | 17.9 |
| Role Play & Narrative | 63.0 | 64.7 | 66.4 |
| Research & Analysis | 52.7 | 63.7 | 74.8 |
Performance by hardware and tool
Every hardware/tool/quantization combination liquidai/lfm2.5-2.6b has been benchmarked on, ranked by peak token generation speed. Last updated August 2026.
| Hardware | Tool | Quant | Best tok/s | Avg tok/s | Memory | Context | Quality | Runs |
|---|---|---|---|---|---|---|---|---|
| NVIDIA GeForce RTX 4070 Ti SUPER | LM Studio | — | 165.9 tok/s | 132.7 tok/s | n/a | 32,768 tokens | 56.8 | 2 |
Benchmark runs
All 2 liquidai/lfm2.5-2.6b runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is lmstudio-community/lfm2.5-2.6b good for coding?
- In our benchmarks, lmstudio-community/lfm2.5-2.6b scores 17.3/100 for coding. It runs at about 132.7 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 lmstudio-community/lfm2.5-2.6b good for agentic (tool-using) tasks?
- In our benchmarks, lmstudio-community/lfm2.5-2.6b scores 81.6/100 for agentic workflows. It runs at about 132.7 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 liquidai/lfm2.5-2.6b for local inference?
- Across 2 community benchmark runs, liquidai/lfm2.5-2.6b reaches up to 165.9 tok/s and averages 132.7 tok/s, with the fastest results on NVIDIA GeForce RTX 4070 Ti SUPER.
- Which tools have been used to run liquidai/lfm2.5-2.6b?
- Benchmarks were submitted using LM Studio. Results are community-contributed and updated as new runs arrive.