Configs › Llama 3.1 70B on Strix Halo: 4.7 tok/s measured
Llama 3.1 70B on Strix Halo: 4.7 tok/s measured
Measured Llama 3.1 70B speeds: 4.7 tok/s on Strix Halo (Q4_K_M, ollama/vulkan). 1 config, each linked to its source, snapshot 2026-10-16.
Snapshot 2026-10-16 · 394 configs on AI boxes and 423 on GPU cards · 123 models. Every number is generated from the tracker snapshot and linked to where it was measured; nothing here is typed by hand.
Quick answers
- How fast is Llama 3.1 70B on Strix Halo?
- 4.7 tok/s single-stream decode (Q4_K_M, ollama/vulkan), the only measured config on AMD Strix Halo (Ryzen AI Max+ 395), measured by hogeheer, source last checked 2026-10-16. source
Measured configs for Llama 3.1 70B
| Model | Hardware | Quant | Backend | Decode | Context | Trust | Source |
|---|---|---|---|---|---|---|---|
| Llama 3.1 70B | Strix Halo | Q4_K_M | ollama/vulkan | 4.7 tok/s (prefill 79.6) | Structured table | source |
More: all configs · hardware · news · contributors · methodology · llms.txt · llms-full.txt · API (OpenAPI) · JSON snapshot. Built by Altronis, private on-prem AI, Singapore.