GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
ademRguez
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
ademRguez
Model Tree
google/gemma-4-E4B
Input Modalities
Output Modalities
Supported Functionality
LoRA adapter (r=16, alpha=32, all attention+MLP projections) + pointer-head for scoring options in typed decisions. Port of open-source Kev training code from Qwen to Gemma-4-E4B. Trained: 30,428 records, 394 optimizer steps, bf16, ~67 min/GPU. Base: google/gemma-4-E4B @411aa17b749aa952df1359d2dcea73917a544d9a.
This does NOT improve Gemma-4-E4B on standard benchmarks. It is a research artifact with measured lower accuracy than baseline Kev-4B. Published to document what did not work.
| Setting | This Model | Kev-4B |
|---|---|---|
| In-domain (600 points) | 0.825 | 0.865 |
| Out-of-domain (600 points) | 0.628 | 0.835 |
| Gold-verified decisions (116 points) | 0.776 | 0.879 |
| Shuffle-consistency in-domain | 0.918 | 0.960 |
| Shuffle-consistency OOD | 0.858 | 0.968 |
Weaknesses: bias toward option A; poor on small option sets (2-opt: 0.64, 3-opt: 0.43); low OOD transfer.
Experiment and evaluation scripts: https://github.com/adem-rguez/Chimera.
Classification-style option selection (choosing among provided options) in-domain with confidence floor (T=0.6112). NOT for open-ended reasoning, planning, math, code, or replacing the base model.
python
from kev.checkpoint import Checkpoint, LoadOptionsimport torch ck = Checkpoint('ademRguez/chimera-ld-gemma-4-e4b-head')tok, model = ck.load('cuda', LoadOptions(dtype=torch.bfloat16)) # Score a decisionstate = "user_text\nthinking_so_far"instruction = "Pick one"options = ["A: first option", "B: second option"]rec = {"state": state, "questions": [{"instr": instruction, "options": options, "label": 0}]}enc = model.encode(tok, rec)probs = model.probs(enc)[0] # softmax scores over optionsRequired: base model at specified revision; install kev from https://github.com/adem-rguez/Chimera.
adapter_model.safetensors, adapter_config.json, head.pt, tokenizer, training_config.json, training_metrics.json.
Kev (Apache-2.0): https://github.com/jaredpalmer/kev Gemma-4-E4B (Apache-2.0 by Google) Training data: decision-v7 suite (upstream Kev project)
If you find this useful, consider buying me a coffee to support continued development:
from kev.checkpoint import Checkpoint, LoadOptionsimport torch ck = Checkpoint('ademRguez/chimera-ld-gemma-4-e4b-head')tok, model = ck.load('cuda', LoadOptions(dtype=torch.bfloat16)) # Score a decisionstate = "user_text\nthinking_so_far"instruction = "Pick one"options = ["A: first option", "B: second option"]rec = {"state": state, "questions": [{"instr": instruction, "options": options, "label": 0}]}enc = model.encode(tok, rec)probs = model.probs(enc)[0] # softmax scores over options