prithivMLmods/Jan-v2-VL-f32-AIO-GGUF

Same weights, converted (fewer layers) vs janhq/Jan-v2-VL-high. prithivMLmods/Jan-v2-VL-f32-AIO-GGUF matches part of janhq/Jan-v2-VL-high (399 of its 750 tensors; the rest of janhq/Jan-v2-VL-high has no counterpart). On the shared part: Same weights stored in another dtype (BF16 → F16/F32).

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

On the Hub · Downloads on Model Pulse