Model benchmarks
zai-org/glm-4.6v-flash local LLM performance
As of August 2026, zai-org/glm-4.6v-flash runs at up to 54.5 tok/s for local inference (best of 4 community benchmark runs across 2 GPUs).
Model size
Unknown
Peak speed
54.5 tok/s
Average speed
23.3 tok/s
Min memory
n/a
Max context
16,384 tokens
Avg output / run
10,742 tokens
Avg runtime / run
9m 47s
Avg quality
36.3
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 | 19.6 | 36.3 | 61.0 |
| Agent Workflow | 37.5 | 68.5 | 84.6 |
| Code Generation | 0.9 | 14.8 | 39.8 |
| Role Play & Narrative | 18.2 | 35.8 | 63.2 |
| Research & Analysis | 11.1 | 26.1 | 57.6 |
Performance by hardware and tool
Every hardware/tool/quantization combination zai-org/glm-4.6v-flash 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 | — | 54.5 tok/s | 54.5 tok/s | n/a | 16,384 tokens | 65.9 | 1 |
| Apple M5 | LM Studio | — | 20.6 tok/s | 20.6 tok/s | n/a | 8,192 tokens | 33.1 | 1 |
| CPU only | LM Studio | — | 13.2 tok/s | 9.1 tok/s | n/a | 1,024 tokens | 23.1 | 2 |
Benchmark runs
All 4 zai-org/glm-4.6v-flash runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is zai-org/glm-4.6v-flash good for coding?
- In our benchmarks, zai-org/glm-4.6v-flash scores 14.8/100 for coding. It runs at about 23.3 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 zai-org/glm-4.6v-flash good for agentic (tool-using) tasks?
- In our benchmarks, zai-org/glm-4.6v-flash scores 68.5/100 for agentic workflows. It runs at about 23.3 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 zai-org/glm-4.6v-flash for local inference?
- Across 4 community benchmark runs, zai-org/glm-4.6v-flash reaches up to 54.5 tok/s and averages 23.3 tok/s, with the fastest results on NVIDIA GeForce RTX 4070 Ti SUPER.
- Which tools have been used to run zai-org/glm-4.6v-flash?
- Benchmarks were submitted using LM Studio. Results are community-contributed and updated as new runs arrive.