Downloads · 30 days
16
31% of all-time downloads
HarmenWessels/Ornith-1.0-9B-int4-symg128-ov
Ornith-1.0-9B-int4-symg128-ov is a image-text-to-text model from HarmenWessels. Use it for the image-text-to-text task on the model card, and read the license before you ship it in a product. The card lists the license as mit.
ornith-ai/Ornith-1.0-9B — an RL post-trained agentic coding model on a Qwen3.5-9B backbone — converted to the OpenVINO™ IR format with weights compressed to INT4 by NNCF.
Downloads · 30 days
16
31% of all-time downloads
All-time downloads
51
Public
Repo size
6.1 GB
Likes
0
Public
Click a slice to open those files.
.bin6.1 GB · 100%
From the Hugging Face model README
ornith-ai/Ornith-1.0-9B — an RL post-trained agentic coding model on a Qwen3.5-9B backbone — converted to the OpenVINO™ IR format with weights compressed to INT4 by NNCF.
optimum-cli export openvino \
--model ornith-ai/Ornith-1.0-9B \
--task image-text-to-text \
--weight-format int4 --sym --group-size 128 --ratio 1.0 \
Ornith-1.0-9B-int4-symg128-ov
PREDEFINED_VISUAL_LM_DATASETS (a raw
corpus raises TypeError: unhashable type: 'list'), contextual is the sole
accepted name, and its image bucket now returns HTTP 403 AllAccessDisabled.
So AWQ / scale-estimation are unreachable for this architecture at present.ACTIVATIONS_SCALE_FACTOR left at the baked 8.0; verified coherent on long
generations (hidden size 4096 does not need the bump larger Gemma/Ministral do).Requires OpenVINO GenAI 2026.3.0.0 or newer (qwen3_5 support). The IR is
VLM-shaped, so it loads through VLMPipeline even for text-only use.
import openvino_genai as ov_genai
pipe = ov_genai.VLMPipeline("Ornith-1.0-9B-int4-symg128-ov", "GPU")
cfg = ov_genai.GenerationConfig()
cfg.max_new_tokens = 512
cfg.do_sample = True
cfg.temperature = 0.6 # vendor-recommended for precise coding
cfg.top_p = 0.95
cfg.top_k = 20
print(pipe.generate("Write a Python function that merges overlapping intervals.",
generation_config=cfg))
It is a reasoning model: output opens with a thinking trace and emits a closing
</think> (the chat template pre-fills the opener), so split on that tag before
showing the answer. Tool calls use the hermes <tool_call> wrapper with a
Qwen3-Coder XML body (<function=name><parameter=k>v</parameter></function>),
not hermes JSON — parsers expecting JSON inside the wrapper will see zero calls.
This model is measured in core-ultra-llm-server — per-task pass/fail (code generation, editing, agentic tool use, analysis, autocomplete) on an Intel Core Ultra 155H iGPU, re-run whenever the engine or server changes. The leaderboard there is authoritative; this card carries no numbers so it cannot go stale. Its benchmark card (decoding, think budget) is cards/HarmenWessels__Ornith-1.0-9B-int4-symg128-ov.yaml.