GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
spiritfather
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
spiritfather
Model Tree
nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16
Input Modalities
Output Modalities
Supported Functionality
| Parameter | Value |
|---|---|
| start_layer_index | 41 |
| end_layer_index | 54 |
| preserve_good_behavior_weight | 0.7358 |
| steer_bad_behavior_weight | 0.1777 |
| 0.9892 |
| neighbor_count | 2 |
| Metric | This model | Original model (nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16) |
|---|---|---|
| Refusals | 6/100 | 100/100 |
| KL divergence | 0.1092 | 0 (by definition) |
Heretic's own evaluation: refusals are Heretic's keyword rule over 100 held-out prompts from mlabonne/harmful_behaviors; KL divergence is first-token KL against the original model over 100 prompts from mlabonne/harmless_alpaca. Thinking off. It is not a safety evaluation and says nothing about what the model will or will not produce.
Load exactly as the original — see NVIDIA's model card for transformers / vLLM / SGLang serving, the reasoning parser and recommended sampling. Everything except the edited weights is unchanged, including trust_remote_code requirements.
Thinking: on by default in the template; pass chat_template_kwargs: {"enable_thinking": false} per request to turn it off.
For llama.cpp, use the GGUF repo (vision via mmproj-*.gguf, MTP via mtp-*.gguf).
end_layer_index is exclusive, so layers 41–53). Before export the trial was restored from its parameters and re-scored; it reproduced its stored 6/100 and 0.1092 exactly.W + B·A computed in fp32 against the original BF16 weights (not the 8-bit search copy) and rounded to BF16 once.language_model.*. Checked before writing that Omni's originals equal the text weights byte for byte, and after writing that the grafted tensors equal the merge; every other tensor and file is copied unchanged.Checked after the build: the vision tower still reads images correctly, and the MTP head still drafts (53% of draft tokens accepted on a llama.cpp Q8_0).
The edit lowers refusals. You decide how to use the model and you answer for it under the laws that apply to you. Intended uses: research, evaluation, red-teaming, creative writing and private deployment. Do not use it to produce illegal content or to target people, or in a public service without your own filtering and moderation.
OpenMDW 1.1, inherited from nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16. NVIDIA's bias.md, explainability.md, privacy.md and safety.md are kept in this repo; they describe the original model.