mradermacher/Q3-8B-GRPO-Balanced-Vision-Freezed-GGUF

Same weights, converted (fewer layers) vs arafid/Q3-8B-GRPO-Balanced-Vision-Freezed. mradermacher/Q3-8B-GRPO-Balanced-Vision-Freezed-GGUF matches part of arafid/Q3-8B-GRPO-Balanced-Vision-Freezed (399 of its 750 tensors; the rest of arafid/Q3-8B-GRPO-Balanced-Vision-Freezed has no counterpart). On the shared part: Same weights stored in another dtype (BF16 → F16/F32).

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

On the Hub · Downloads on Model Pulse