DeepSeek-V4-Flash is a preview of the DeepSeek-V4 series, a Mixture-of-Experts model with 284B total parameters and 13B activated, built for efficient reasoning across a 1M-token context window.
56.7K Pulls 1 Tag Updated 2 weeks ago
DeepSeek-V4-Pro is a frontier Mixture-of-Experts model with a 1M-token context window and three reasoning modes.
47.2K Pulls 1 Tag Updated 1 week ago
An open-source Mixture-of-Experts code language model that achieves performance comparable to GPT4-Turbo in code-specific tasks.
2.5M Pulls 64 Tags Updated 1 year ago
106 Pulls 1 Tag Updated 2 weeks ago
DeepSeek-V3-Pruned-Coder-411B is a pruned version of the DeepSeek-V3 reduced from 256 experts to 160 experts, The pruned model is mainly used for code generation.
1,371 Pulls 5 Tags Updated 1 year ago
DeepSeep V3 from March 2025 Merged from Unsloth's HF - 671B params - Q8_0/713 GB & Q4_K_M/404 GB
952 Pulls 4 Tags Updated 1 year ago
This is a brand new Mixture of Export (MoE) model from DeepSeek, specializing in coding instructions. (quantized IQ4_XS)
11.1K Pulls 3 Tags Updated 3 months ago
6,452 Pulls 23 Tags Updated 1 year ago
This model has been developed based on DistilQwen2.5-DS3-0324-Series.
1,175 Pulls 7 Tags Updated 1 year ago
deepseek-v3-0324-Quants. - Q2_K is the lowest here - quantized = round((original - zero_point) / scale)
1,115 Pulls 1 Tag Updated 1 year ago
This model was converted to GGUF format from deepseek-ai/DeepSeek-Coder-V2-Lite-Base using llama.cpp via the ggml.ai's GGUF-my-repo space. Refer to the original model card for more details on the model.
1,021 Pulls 1 Tag Updated 1 year ago
dynamic quants from unsloth, merged
293 Pulls 1 Tag Updated 1 year ago
179 Pulls 1 Tag Updated 2 years ago
A strong Mixture-of-Experts (MoE) language model with 671B total parameters with 37B activated for each token.
21 Pulls 1 Tag Updated 1 year ago
Merged Unsloth's Dynamic Quantization
1,380 Pulls 1 Tag Updated 1 year ago
Latest DeepSeek_V3 model Q4
249 Pulls 1 Tag Updated 1 year ago