mirror of
https://github.com/kvcache-ai/ktransformers.git
synced 2025-09-05 03:59:54 +00:00
5ec33d046d
use marlin for lm_head, lm_head only calc last token for prefill extend context window to 19K for DeepSeek-V3/R1 within 24GB VRAM |
||
|---|---|---|
| .. | ||
| DeepSeek-V2-Chat-multi-gpu-4.yaml | ||
| DeepSeek-V2-Chat-multi-gpu.yaml | ||
| DeepSeek-V2-Chat.yaml | ||
| DeepSeek-V2-Lite-Chat-multi-gpu.yaml | ||
| DeepSeek-V2-Lite-Chat.yaml | ||
| DeepSeek-V3-Chat-multi-gpu-4.yaml | ||
| DeepSeek-V3-Chat-multi-gpu-8.yaml | ||
| DeepSeek-V3-Chat-multi-gpu-marlin.yaml | ||
| DeepSeek-V3-Chat-multi-gpu.yaml | ||
| DeepSeek-V3-Chat.yaml | ||
| Internlm2_5-7b-Chat-1m.yaml | ||
| Mixtral.yaml | ||
| Qwen2-57B-A14B-Instruct-multi-gpu.yaml | ||
| Qwen2-57B-A14B-Instruct.yaml | ||