English | 📖 中文文档
Heretic Trial 63 BF16 merge · compatible MTP head · APEX GGUFs + mmproj
🐦 About this releaseOrnith-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.
🧠 Model Details| Architecture | Qwen3.5 MoE, multimodal |
| Parameters | 35B total, 3B active per token · 256 experts, 8 active |
| Layers | 40 transformer layers + 1 MTP layer (785 BF16 tensors) |
| Context | 262,144 tokens |
| Weights | BF16 safetensors, 17 shards; audited LoRA merge |
| License | MIT |
Selected parameters: direction_index=21.66; attention o_proj range 1.55–2.05; MLP down_proj range 1.64–1.84.
| Refusal prescreen / full evaluation | 2 / 30 · 11 / 100 |
| Validation KL mean / p95 / max | 0.0242 / 0.1477 / 0.1648 |
| Generation health | Passed · top-1 disagreement 2.0% |
| Mode | ToolCall | BugFind | HermesAgent | Max / Effective |
|---|---|---|---|---|
| Thinking ↗ | 100 | 98 | 89 | 95.0 / 90.0 |
| No Thinking ↗ | 97 | 97 | 85 | 92.2 / 76.2 |
Latest result cards: Thinking has 5 retry-penalty points; No Thinking has 16.
🚀 Usagellama.cpp (vision + text)
hf download SC117/Ornith-1.0-35B-Heretic-MTP-APEX-GGUF --include "*.gguf" --local-dir ./models ./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
🎛️ Recommended Settings| General / Coding | temperature=0.6, top_p=0.95, top_k=20, do_sample=true |
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.
📦 APEX Quantization Tiers| File | Size | Best For |
|---|---|---|
*-APEX-I-Quality.gguf | 21.87 GiB | Highest quality |
*-APEX-I-Balanced.gguf | 24.33 GiB | Best all-rounder |
*-APEX-I-Compact.gguf | 16.14 GiB | Best quality/size ratio |
*-APEX-I-Mini.gguf | 13.29 GiB | Most compact |
@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}
}
model-00006-of-00017.safetensors 5.0 GBmodel-00009-of-00017.safetensors 5.0 GBmodel-00012-of-00017.safetensors 5.0 GBmodel-00015-of-00017.safetensors 5.0 GBmodel-00003-of-00017.safetensors 5.0 GBmodel-00008-of-00017.safetensors 4.5 GBmodel-00011-of-00017.safetensors 4.5 GBmodel-00014-of-00017.safetensors 4.5 GBmodel-00005-of-00017.safetensors 4.5 GBmodel-00002-of-00017.safetensors 4.5 GBmodel-00001-of-00017.safetensors 4.3 GBmodel-00007-of-00017.safetensors 4.0 GBmodel-00010-of-00017.safetensors 4.0 GBmodel-00013-of-00017.safetensors 4.0 GBmodel-00004-of-00017.safetensors 4.0 GBmodel-00016-of-00017.safetensors 2.6 GBmodel-00017-of-00017.safetensors 1.7 GBornith-ai/Ornith-1.0-35B