Configs › DeepSeek V4 Flash 284B on Strix Halo: 32 tok/s measured
DeepSeek V4 Flash 284B on Strix Halo: 32 tok/s measured
Measured DeepSeek V4 Flash 284B speeds: 32 tok/s on Strix Halo (IQ2_XXS, speculative decode, mtp). 2 configs, 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 DeepSeek V4 Flash 284B on Strix Halo?
- 32 tok/s single-stream decode (IQ2_XXS, speculative decode, mtp), the fastest of 2 measured configs on AMD Strix Halo (Ryzen AI Max+ 395), measured by julianmb, source last checked 2026-09-16. source
Measured configs for DeepSeek V4 Flash 284B
| Model | Hardware | Quant | Backend | Decode | Context | Trust | Source |
|---|---|---|---|---|---|---|---|
| DeepSeek V4 Flash 284B | Strix Halo | IQ2_XXS | speculative decode, mtp | 32 tok/s | Owner-submitted | source | |
| DeepSeek V4 Flash 284B | Strix Halo | UD-IQ2_XXS | llama.cpp/vulkan | 13.3 tok/s (prefill 155.6) | 512 | 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.