Tesslate/Tessa-T1-7B
LoRA-style vs Qwen/Qwen2.5-Coder-7B-Instruct. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, gate_proj, k_proj, o_proj, q_proj, up_proj, v_proj changed, in 28 of 28 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/57, MLP 84/84, Attention 112/196, lm_head 0/1
5 derivatives checked
- bartowski/Tesslate_Tessa-T1-7B-GGUF: Same weights, converted
- second-state/Tessa-T1-7B-GGUF: Same weights, converted
- csabakecskemeti/Tesslate.Tessa-T1-7B-GGUF: Same weights, converted
- gaianet/Tessa-T1-7B-GGUF: Same weights, converted
- mradermacher/Tessa-T1-7B-GGUF: Same weights, converted
1 other repos hold byte-for-byte identical weight files.