19 models
DeepSeek-V4-Pro is a frontier Mixture-of-Experts model with a large context window and three reasoning modes.
Jackrong/Qwen3.5-9B-DeepSeek-V4-Flash - is an efficient reasoning model distilled using high-quality data from DeepSeek-V4. + vision. ollama v.0.30.0-rc20 +
pull from hf.co/Jackrong/Qwen3.5-9B-DeepSeek-V4-Flash-GGUF
Jackrong/Qwen3.5-9B-DeepSeek-V4-Flash - is an efficient reasoning model distilled using high-quality data from DeepSeek-V4.
DeepSeek-V4-Flash 284B MoE (13B active), single-file IQ2_M merged from AtomicChat's GGUF shards so Ollama can pull it. ~33 tok/s on a 128 GB Apple Silicon Mac. Model by DeepSeek (MIT), quantization by AtomicChat.
No Oficial 29 t/s en 8 VRAM - Calidad a escala fronteriza como GPT-5.5, DeepSeek-V4-pro y Kimi-K2.6.
To run DeepSeek-V4-Flash in full precision lossless, run Q4 (UD-Q4_K_XL), It is 155GB. vision tools thinking
DeepSeek-V4-Flash-Fast is a custom agent build based on DeepSeek-V4-Flash (Open Weights, Apache-2.0, MoE, ~285B total / ~20B active). Weights in low-bit quantization for fully CPU-only deployment:
To run DeepSeek-V4-Flash in full precision lossless, run IQ1 (UD-IQ1_S), It is 82GB. vision tools thinking
ollama run hf.co/unsloth/DeepSeek-V4-Flash-GGUF:UD-Q4_K_XL
To run DeepSeek-V4-Flash in full precision lossless, run Q3 (UD-Q3_K_M), It is 129 GB. vision tools thinking
A customized DeepSeek-V4-Pro
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.
DeepSeep V3 from March 2025 Merged from Unsloth's HF - 671B params - Q8_0/713 GB & Q4_K_M/404 GB
Uncensored DeepSeek-R1-0528-Qwen3-8B via Heretic: 4/100 refusals (from 89), KL 0.041. Reasoning model, MIT.
Merged Unsloth's Dynamic Quantization
Latest DeepSeek_V3 model Q4