DeepSeek-OCR 2 BASE

No machine on this network has measured this model yet. Everything below is what the catalog knows exists — where the weights come from, how large each build is, and whose work it is. Nothing here is a claim about what the model can do; that only appears once a machine has run it and signed the result.

hf.co/SandLogicTechnologies/DeepSeek-OCR-2-GGUF:IQ3_Mollama · IQ3_M · 1.3 GiB · quantizer (SandLogicTechnologies)

hf.co/SandLogicTechnologies/DeepSeek-OCR-2-GGUF:IQ4_XSollama · IQ4_XS · 1.5 GiB · quantizer (SandLogicTechnologies)

hf.co/sabafallah/DeepSeek-OCR-2-GGUF:Q4_K_Mollama · Q4_K_M · 1.8 GiB · quantizer (sabafallah)

hf.co/sabafallah/DeepSeek-OCR-2-GGUF:Q8_0ollama · Q8_0 · 2.9 GiB · quantizer (sabafallah)

richarddavison/DeepSeek-OCR-2-FP8vllm · FP8 · 3.5 GiB · quantizer (richarddavison)

bartsolutions/DeepSeek-OCR-2-GPTQ-INT8vllm · GPTQ · 4.2 GiB · quantizer (bartsolutions)

hf.co/sabafallah/DeepSeek-OCR-2-GGUF:BF16ollama · BF16 · 5.5 GiB · quantizer (sabafallah)

deepseek-ai/DeepSeek-OCR-2vllm · BF16 · 6.3 GiB · vendor (deepseek-ai)

derivatives — somebody else's change to these weights

Fine-tunes, merges and abliterations of this member, each its own model with its own page. None of their numbers are counted above.

DeepSeek-OCR-2-Q4_K_M.ggufnot measured DeepSeek-OCR-2not measured AX-DeepSeek-OCR-2-MLX-AXQ-MXFP8not measured AX-DeepSeek-OCR-2-CUDA-AXQ-NVFP4-W4A16not measured DeepSeek-OCR-2not measured Universal-DeepSeek-OCR-2not measured DeepSeek-OCR-2-Q4_K_M-GGUFnot measured Arabic-OCR-DeepSeek-OCR-2not measured DeepSeek-OCR-2-8bitnot measured DeepSeek-OCR-2not measured DeepSeek-OCR-2not measured