Model benchmarks
glm-4.7-flash:q4_K_M local LLM performance
As of November 2026, glm-4.7-flash:q4_K_M runs at up to 26.2 tok/s for local inference (best of 2 community benchmark runs across 1 GPU).
Model size
29.9B
Peak speed
26.2 tok/s
Average speed
23.6 tok/s
Avg PP
269.4 tok/s
Min memory
32.4 GB
Max context
65,536 tokens
Avg output / run
15,672 tokens
Avg runtime / run
11m 56s
Avg quality
65.8
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 | 64.0 | 65.8 | 67.6 |
| Agent Workflow | 62.5 | 64.1 | 65.7 |
| Code Generation | 54.0 | 59.5 | 65.1 |
| Role Play & Narrative | 66.7 | 69.5 | 72.3 |
| Research & Analysis | 67.0 | 69.9 | 72.7 |
Performance by hardware and tool
Every hardware/tool/quantization combination glm-4.7-flash:q4_K_M has been benchmarked on, ranked by peak token generation speed. Last updated November 2026.
| Hardware | Tool | Quant | Best tok/s | Avg tok/s | Memory | Context | Quality | Runs |
|---|---|---|---|---|---|---|---|---|
| NVIDIA RTX 2000 Ada Generation Laptop GPU | Ollama | Q4_K_M | 26.2 tok/s | 23.6 tok/s | 32.4 GB | 65,536 tokens | 65.8 | 2 |
Benchmark runs
All 2 glm-4.7-flash:q4_K_M runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is glm-4.7-flash:q4_K_M good for coding?
- In our benchmarks, glm-4.7-flash:q4_K_M scores 59.5/100 for coding. It runs at about 23.6 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 glm-4.7-flash:q4_K_M good for agentic (tool-using) tasks?
- In our benchmarks, glm-4.7-flash:q4_K_M scores 64.1/100 for agentic workflows. It runs at about 23.6 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 glm-4.7-flash:q4_K_M for local inference?
- Across 2 community benchmark runs, glm-4.7-flash:q4_K_M reaches up to 26.2 tok/s and averages 23.6 tok/s, with the fastest results on NVIDIA RTX 2000 Ada Generation Laptop GPU.
- How much memory does glm-4.7-flash:q4_K_M need?
- The leanest observed configuration used about 32.4 GB of memory (quantizations tested: Q4_K_M).
- Which tools have been used to run glm-4.7-flash:q4_K_M?
- Benchmarks were submitted using Ollama. Results are community-contributed and updated as new runs arrive.