canada-quant
/

DeepSeek-V4-Flash-W4A16-FP8

mixture-of-experts

Mixture of Experts

compressed-tensors

Model card Files Files and versions

DeepSeek-V4-Flash-W4A16-FP8 / tokenizer.json

pastapaul's picture

Phase 3b: AWQ-W4A16 quantization (FP8_BLOCK attn + W4A16 routed experts)

2e7ef6a verified 20 days ago

history contribute delete

10.1 MB

File too large to display, you can check the raw version instead.