Model benchmarks
smollm2-135m-instruct local LLM performance
As of October 2026, smollm2-135m-instruct runs at up to 109.5 tok/s for local inference (best of 2 community benchmark runs across 1 GPU).
Model size
135M
Peak speed
109.5 tok/s
Average speed
107.7 tok/s
Avg PP
3990.5 tok/s
Min memory
531 MB
Max context
16,384 tokens
Avg output / run
3,242 tokens
Avg runtime / run
35s
Avg quality
10.0
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 | 9.5 | 10.0 | 10.4 |
| Agent Workflow | 9.9 | 13.5 | 17.0 |
| Code Generation | 0.1 | 0.8 | 1.4 |
| Role Play & Narrative | 18.7 | 19.2 | 19.6 |
| Research & Analysis | 5.8 | 6.6 | 7.3 |
Performance by hardware and tool
Every hardware/tool/quantization combination smollm2-135m-instruct 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 |
|---|---|---|---|---|---|---|---|---|
| AMD Radeon Vega Series / Radeon Vega Mobile Series | Bionic | — | 109.5 tok/s | 107.7 tok/s | 531 MB | 16,384 tokens | 10.0 | 2 |
Benchmark runs
All 2 smollm2-135m-instruct runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is smollm2-135m-instruct good for agentic (tool-using) tasks?
- In our benchmarks, smollm2-135m-instruct scores 13.5/100 for agentic workflows. It runs at about 107.7 tok/s, so if you want more speed, qwen3.8-flash-next-q2_0 is faster (~127.9 tok/s) and still scores well for agentic workflows (85.6/100).
- How fast is smollm2-135m-instruct for local inference?
- Across 2 community benchmark runs, smollm2-135m-instruct reaches up to 109.5 tok/s and averages 107.7 tok/s, with the fastest results on AMD Radeon Vega Series / Radeon Vega Mobile Series.
- How much memory does smollm2-135m-instruct need?
- The leanest observed configuration used about 531 MB of memory.
- Which tools have been used to run smollm2-135m-instruct?
- Benchmarks were submitted using Bionic. Results are community-contributed and updated as new runs arrive.