mradermacher/sphinx-qwen3-8b-GGUF

Same weights, converted (fewer layers) vs maveryn/sphinx-qwen3-8b. mradermacher/sphinx-qwen3-8b-GGUF matches part of maveryn/sphinx-qwen3-8b (399 of its 750 tensors; the rest of maveryn/sphinx-qwen3-8b has no counterpart). On the shared part: Same weights stored in another dtype (BF16 → F16/F32).

Tensors changed: MLP 0/108, Norms 0/145, Attention 0/144, lm_head 0/1, Embeddings 0/1

On the Hub · Downloads on Model Pulse