Model benchmarks
LFM2.5-24B-A2B.APEX-I-MiniPlus-V2.1 local LLM performance
As of October 2026, LFM2.5-24B-A2B.APEX-I-MiniPlus-V2.1 runs at up to 29.6 tok/s for local inference (best of 3 community benchmark runs across 1 GPU).
Model size
24B
Peak speed
29.6 tok/s
Average speed
24.8 tok/s
Avg PP
116.9 tok/s
Min memory
n/a
Max context
65,536 tokens
Avg output / run
27,836 tokens
Avg runtime / run
21m 42s
Avg quality
65.1
Benchmark runs
3
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 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 | 57.8 | 65.1 | 70.2 |
| Agent Workflow | 13.7 | 45.5 | 65.2 |
| Code Generation | 58.1 | 61.7 | 64.7 |
| Role Play & Narrative | 71.3 | 75.6 | 81.2 |
| Research & Analysis | 72.2 | 77.6 | 82.3 |
Performance by hardware and tool
Every hardware/tool/quantization combination LFM2.5-24B-A2B.APEX-I-MiniPlus-V2.1 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 Graphics | llama.cpp | — | 29.6 tok/s | 29.6 tok/s | n/a | 65,536 tokens | 70.4 | 1 |
| Intel Graphics | llama.cpp:vulkan | — | 28.4 tok/s | 28.4 tok/s | n/a | 65,536 tokens | 56.6 | 1 |
| Intel Graphics | llama.cpp:sycl | — | 16.3 tok/s | 16.3 tok/s | n/a | 65,536 tokens | 68.2 | 1 |
Benchmark runs
All 3 LFM2.5-24B-A2B.APEX-I-MiniPlus-V2.1 runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is LFM2.5-24B-A2B.APEX-I-MiniPlus-V2.1 good for coding?
- In our benchmarks, LFM2.5-24B-A2B.APEX-I-MiniPlus-V2.1 scores 61.7/100 for coding. It runs at about 24.8 tok/s, so if you want more speed, local-qwen is faster (~60.2 tok/s) and still scores well for coding (84.8/100).
- Is LFM2.5-24B-A2B.APEX-I-MiniPlus-V2.1 good for agentic (tool-using) tasks?
- In our benchmarks, LFM2.5-24B-A2B.APEX-I-MiniPlus-V2.1 scores 45.5/100 for agentic workflows. It runs at about 24.8 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 LFM2.5-24B-A2B.APEX-I-MiniPlus-V2.1 for local inference?
- Across 3 community benchmark runs, LFM2.5-24B-A2B.APEX-I-MiniPlus-V2.1 reaches up to 29.6 tok/s and averages 24.8 tok/s, with the fastest results on Intel Graphics.
- Which tools have been used to run LFM2.5-24B-A2B.APEX-I-MiniPlus-V2.1?
- Benchmarks were submitted using llama.cpp, llama.cpp:sycl, llama.cpp:vulkan. Results are community-contributed and updated as new runs arrive.