Model benchmarks
Ling-3.0-tiny-oQ8e local LLM performance
As of September 2026, Ling-3.0-tiny-oQ8e runs at up to 137.8 tok/s for local inference (best of 4 community benchmark runs across 2 GPUs).
Model size
Unknown
Peak speed
137.8 tok/s
Average speed
124.9 tok/s
Avg prefill
1826.4 tok/s
Min memory
8.3 GB
Max context
131,072 tokens
Avg output / run
73,086 tokens
Avg runtime / run
9m 14s
Avg quality
60.1
Benchmark runs
4
GPUs tested
2
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 4 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 | 56.1 | 60.1 | 64.2 |
| Agent Workflow | 46.4 | 72.3 | 84.8 |
| Code Generation | 38.5 | 47.3 | 56.8 |
| Role Play & Narrative | 35.7 | 45.5 | 50.9 |
| Research & Analysis | 68.9 | 75.2 | 82.4 |
Performance by hardware and tool
Every hardware/tool/quantization combination Ling-3.0-tiny-oQ8e has been benchmarked on, ranked by peak token generation speed. Last updated September 2026.
| Hardware | Tool | Quant | Best tok/s | Avg tok/s | Memory | Context | Quality | Runs |
|---|---|---|---|---|---|---|---|---|
| Apple M4 Max | oMLX | — | 137.8 tok/s | 126.0 tok/s | 8.6 GB | 131,072 tokens | 58.7 | 3 |
| Apple M5 Max | oMLX | — | 121.6 tok/s | 121.6 tok/s | 8.3 GB | 131,072 tokens | 64.3 | 1 |
Frequently asked questions
- Is Ling-3.0-tiny-oQ8e good for coding?
- In our benchmarks, Ling-3.0-tiny-oQ8e scores 47.3/100 for coding. It runs at about 124.9 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 Ling-3.0-tiny-oQ8e good for agentic (tool-using) tasks?
- In our benchmarks, Ling-3.0-tiny-oQ8e scores 72.3/100 for agentic workflows. It runs at about 124.9 tok/s, so if you want more speed, Tiel-Coder-35B-A3B-Q4_K_S is faster (~165.0 tok/s) and still scores well for agentic workflows (89.3/100).
- How fast is Ling-3.0-tiny-oQ8e for local inference?
- Across 4 community benchmark runs, Ling-3.0-tiny-oQ8e reaches up to 137.8 tok/s and averages 124.9 tok/s, with the fastest results on Apple M4 Max.
- How much memory does Ling-3.0-tiny-oQ8e need?
- The leanest observed configuration used about 8.3 GB of memory.
- Which tools have been used to run Ling-3.0-tiny-oQ8e?
- Benchmarks were submitted using oMLX. Results are community-contributed and updated as new runs arrive.