No official torrent yet β requesting it tells the flock what to host next. You cannot seed a catalog-only page.
Cataloged from SC117/Ornith-1.0-35B-Heretic-MTP-APEX-GGUF
All models by SC117 βEnglish | π δΈζζζ‘£
Heretic Trial 63 BF16 merge Β· compatible MTP head Β· APEX GGUFs + mmproj
Ornith-1.0-35B is an agentic coding model from DeepReinforce AI, based on Qwen3.5 MoE. This experimental derivative merges the selected Heretic 1.4.0 Trial 63 LoRA into BF16 weights.
It also includes a compatible Qwen3.5 MTP head, a BF16 vision projector, and APEX mixed-precision GGUFs. Refusal-modified models can behave differently from the original; evaluate before deployment. License: MIT.
Selected parameters: direction_index=21.66; attention o_proj range 1.55β2.05; MLP down_proj range 1.64β1.84.
Latest result cards: Thinking has 5 retry-penalty points; No Thinking has 16.
llama.cpp (vision + text + MTP acceleration)
This model preserves the native MTP head. To enable MTP speculative decoding, add:
Example:
The optimal --spec-draft-n-max value depends on hardware and MTP acceptance rate.
Without these parameters, the model will work normally but will not use MTP acceleration.
These GGUFs use APEX, an MoE-aware mixed-precision quantization method. It allocates precision according to tensor role and layer position; the MTP layer remains Q8_0 and the matching vision projector remains BF16.
--spec-type draft-mtp --spec-draft-n-max 2./llama-server \-m ./models/Ornith-1.0-35B-Heretic-MTP-APEX-I-Compact.gguf \--mmproj ./models/mmproj-Ornith-1.0-35B-Heretic-MTP-BF16.gguf \-ngl 99 \-c 131072 \--spec-type draft-mtp \--spec-draft-n-max 2@misc{ornith-35b, title = {{Ornith-1.0-35B}: Agentic Coding, Open to All}, url = {https://deep-reinforce.com/ornith_1_0.html}, author = {{DeepReinforce Team}}, year = {2026}}