Atream 5ec33d046d optimize gguf dequant, save mem, support Q2_K ...

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

2025-02-22 06:13:01 +00:00

..

DeepSeek-V2-Chat-multi-gpu-4.yaml

optimize gguf dequant, save mem, support Q2_K

2025-02-22 06:13:01 +00:00

DeepSeek-V2-Chat-multi-gpu.yaml

optimize gguf dequant, save mem, support Q2_K

2025-02-22 06:13:01 +00:00

DeepSeek-V2-Chat.yaml

optimize gguf dequant, save mem, support Q2_K

2025-02-22 06:13:01 +00:00

DeepSeek-V2-Lite-Chat-multi-gpu.yaml

optimize gguf dequant, save mem, support Q2_K

2025-02-22 06:13:01 +00:00

DeepSeek-V2-Lite-Chat.yaml

optimize gguf dequant, save mem, support Q2_K

2025-02-22 06:13:01 +00:00

DeepSeek-V3-Chat-multi-gpu-4.yaml

optimize gguf dequant, save mem, support Q2_K

2025-02-22 06:13:01 +00:00

DeepSeek-V3-Chat-multi-gpu-8.yaml

optimize gguf dequant, save mem, support Q2_K

2025-02-22 06:13:01 +00:00

DeepSeek-V3-Chat-multi-gpu-marlin.yaml

optimize gguf dequant, save mem, support Q2_K

2025-02-22 06:13:01 +00:00

DeepSeek-V3-Chat-multi-gpu.yaml

optimize gguf dequant, save mem, support Q2_K

2025-02-22 06:13:01 +00:00

DeepSeek-V3-Chat.yaml

optimize GPU

2025-02-21 05:06:57 +00:00

Internlm2_5-7b-Chat-1m.yaml

[feature] release 0.1.3

2024-08-28 16:11:43 +00:00

Mixtral.yaml

optimize gguf dequant, save mem, support Q2_K

2025-02-22 06:13:01 +00:00

Qwen2-57B-A14B-Instruct-multi-gpu.yaml

optimize gguf dequant, save mem, support Q2_K

2025-02-22 06:13:01 +00:00

Qwen2-57B-A14B-Instruct.yaml

optimize gguf dequant, save mem, support Q2_K

2025-02-22 06:13:01 +00:00