DeepSeek
DeepSeek v3
- Weights at 16-bit
- 1,342 GB
- Weights at 8-bit (Q8_0)
- 713 GB
- Weights at 4-bit (Q4_K_M)
- 407 GB
- Smallest card at 4-bit
- More than one card
- Parameters
- Active parameters
- Layers
- KV heads
- Context
- Licence
Specs
- Active parameters
- Attention heads
- Context
- Downloads (30 days)
- Experts
- Experts per token
- Family
- GGUF downloads
-
repo bytes files quant unsloth/DeepSeek-V3-GGUF 221,253,686,368 5 entries Q2_K_XS unsloth/DeepSeek-V3-GGUF 244,245,533,792 5 entries Q2_K_L unsloth/DeepSeek-V3-GGUF 319,198,395,296 7 entries Q3_K_M unsloth/DeepSeek-V3-GGUF 404,430,187,072 9 entries Q4_K_M unsloth/DeepSeek-V3-GGUF 475,396,557,600 10 entries Q5_K_M unsloth/DeepSeek-V3-GGUF 550,798,326,368 12 entries Q6_K unsloth/DeepSeek-V3-GGUF 713,286,514,080 16 entries Q8_0 - Hidden size
- KV heads
- KV latent rank (MLA)
- Layers
- Licence
- Parameters
- Parameters by type
- Publisher
- Released
- RoPE key size (MLA)
- Shared experts
- Task
More from DeepSeek
- DeepSeek deepseek-coder-1.3b-instruct
- DeepSeek deepseek-coder-6.7b-base
- DeepSeek deepseek-coder-6.7b-instruct
- DeepSeek deepseek-coder-7b-instruct-v1.5
- DeepSeek DeepSeek-Coder-V2-Lite-Instruct
- DeepSeek deepseek-moe-16b-base
- DeepSeek deepseek-moe-16b-chat
- DeepSeek DeepSeek-R1
- DeepSeek DeepSeek-R1-0528
- DeepSeek DeepSeek-R1-0528-Qwen3-8B
- DeepSeek DeepSeek-R1-Distill-Llama-8B
- DeepSeek DeepSeek-R1-Distill-Llama-70B