Dimensity 9400
11 records across 2 models (quants: Q4_K_M, Q8_0).
| hardware | model | quant | backend | tok/s | pp tok/s | W | ttft s | ctx/tg | date | class | flags | source |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Dimensity 9400 | Qwen3-Embedding-0.6B | Q8_0 | llama.cpp (CPU) | – | 102.33 | – | – | 512 | 2026-10-03 | community | – | source |
| Dimensity 9400 | Qwen3-Embedding-0.6B | Q8_0 | llama.cpp (CPU) | – | 240.05 | – | – | 512 | 2026-10-03 | community | – | source |
| Dimensity 9400 | Qwen3-Embedding-0.6B | Q8_0 | llama.cpp (CPU) | – | 229.56 | – | – | 512 | 2026-10-03 | community | – | source |
| Dimensity 9400 | Qwen3-Embedding-0.6B | Q8_0 | llama.cpp (CPU) | – | 259.77 | – | – | 512 | 2026-10-03 | community | – | source |
| Dimensity 9400 | Qwen3-Embedding-0.6B | Q8_0 | llama.cpp (CPU) | – | 123.45 | – | – | 512 | 2026-10-03 | community | – | source |
| Dimensity 9400 | Qwen3.5-0.8B 752M | Q4_K_M | llama.cpp (Vulkan) | 36.88 | 54.04 | – | – | tg32 | 2026-10-05 | community | – | source |
| Dimensity 9400 | Qwen3.5-0.8B 752M | Q4_K_M | llama.cpp (Vulkan) | 36.62 | 125.78 | – | – | tg32 | 2026-10-05 | community | – | source |
| Dimensity 9400 | Qwen3.5-0.8B 752M | Q4_K_M | llama.cpp (Vulkan) | 36.4 | 53.3 | – | – | tg32 | 2026-10-05 | community | – | source |
| Dimensity 9400 | Qwen3.5-0.8B 752M | Q4_K_M | llama.cpp (Vulkan) | 36.03 | 54.53 | – | – | tg32 | 2026-10-05 | community | – | source |
| Dimensity 9400 | Qwen3.5-0.8B 752M | Q4_K_M | llama.cpp (Vulkan) | 35.8 | 117.27 | – | – | tg32 | 2026-10-05 | community | – | source |
| Dimensity 9400 | Qwen3.5-0.8B 752M | Q4_K_M | llama.cpp (CPU) | 2.95 | 51.19 | – | – | tg32 | 2026-10-05 | community | – | source |
All rows cite their source. Provenance: community 40018 sourced 1085