ISTNetworks/ultravox-v0_6-qwen-3-32b-arabic
Same weights, repacked (fewer layers) vs Qwen/Qwen3-32B. Qwen/Qwen3-32B matches part of ISTNetworks/ultravox-v0_6-qwen-3-32b-arabic (707 of its 1198 tensors; the rest of ISTNetworks/ultravox-v0_6-qwen-3-32b-arabic has no counterpart). On the shared part: Same weights in different files (re-saved, re-sharded or new metadata), plus 491 new tensors.
Tensors changed: Embeddings 0/1, Norms 0/257, MLP 0/192, Attention 0/256, lm_head 0/1
1 derivatives checked
- mradermacher/ultravox-v0_6-qwen-3-32b-arabic-GGUF: Quantization (fewer layers)