GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
dknguyen2304
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
dknguyen2304
Model Tree
Qwen/Qwen3-4B
Input Modalities
Output Modalities
Supported Functionality
This model is trained to produce typed decisions with calibrated confidence — given a short piece of text and a small set of labeled options (2-20), it outputs a single-line JSON object naming its chosen label and a confidence score for that choice:
json
{"answer": "B", "confidence": 0.94}No chain-of-thought, no free-form generation — one short structured decision per input, matching a "System One" inference style rather than a reasoning model.
This is an independent, non-affiliated reproduction of the training methodology
described by TypeSafe AI's Jev model (RLCD instead of plain RLHF /
correctness-only reward). It is not TypeSafe's Jev model, not derived from their weights, and
not endorsed by them — named openjev (not jev) for exactly this reason, mirroring why the
original open-source clone this project descends from
(workszop/openjev) was itself renamed to SemIf to
avoid trademark confusion with the commercial product.
A binary correct/incorrect reward (the default in most GRPO recipes, e.g. verl's own GSM8K example) only trains for accuracy — a model can learn to always claim high confidence and still score well on average. RLCD instead scores every rollout with a Brier score proper scoring rule:
text
reward = 1 - (confidence - is_correct)^2This is maximized only when the model is both correct and confident, and penalizes confident-but-wrong answers harder than unsure-but-wrong ones — the calibration property that lets downstream software act automatically on high-confidence decisions and route low-confidence ones to human review, which is the whole point of a typed-decision model.
| Base model | Qwen/Qwen3-4B |
| Algorithm | GRPO (algorithm.adv_estimator=grpo) |
| Framework | XiaomiMiMo/verl 0.9.0.dev, FSDP + vLLM rollout |
| Hardware | 2x NVIDIA H100-80GB, single node |
| Response length | 64 tokens (short structured JSON, no CoT) |
| Thinking mode | disabled (apply_chat_template_kwargs.enable_thinking=false) |
| Reward | custom RLCD/Brier score (reward.custom_reward_function) |
Important inference note: Qwen3's chat template defaults to "thinking mode." This model
was trained with it explicitly disabled — at inference time you must pass the same
template kwarg (e.g. chat_template_kwargs={"enable_thinking": false} on an OpenAI-compatible
vLLM server, or enable_thinking=False to tokenizer.apply_chat_template), otherwise the model
will spend its whole (short) response budget on <think>...</think> reasoning and never emit
the JSON decision.
python
from transformers import AutoModelForCausalLM, AutoTokenizer model = AutoModelForCausalLM.from_pretrained("dknguyen2304/openjev-rlcd-qwen3-4b-v1")tokenizer = AutoTokenizer.from_pretrained("dknguyen2304/openjev-rlcd-qwen3-4b-v1") prompt = tokenizer.apply_chat_template( [{"role": "user", "content": ( "You are a typed-decision engine (System One). Given the input, choose exactly ONE " "label from the options below and report your confidence that this label is correct.\n" "Respond with ONLY a single-line JSON object in this exact format, no other text, no " 'reasoning, no markdown fences:\n{"answer": "<LETTER>", "confidence": <number between 0 ' "and 1>}\n\n" "Input: add this song to my road trip playlist\n\n" "Options:\nA) BookRestaurant\nB) AddToPlaylist\nC) GetWeather\nD) PlayMusic\n" "E) RateBook\nF) SearchCreativeWork\nG) SearchScreeningEvent" )}], tokenize=False, add_generation_prompt=True, enable_thinking=False,){"answer": "B", "confidence": 0.94}from transformers import AutoModelForCausalLM, AutoTokenizer model = AutoModelForCausalLM.from_pretrained("dknguyen2304/openjev-rlcd-qwen3-4b-v1")tokenizer = AutoTokenizer.from_pretrained("dknguyen2304/openjev-rlcd-qwen3-4b-v1") prompt = tokenizer.apply_chat_template( [{"role": "user", "content": ( "You are a typed-decision engine (System One). Given the input, choose exactly ONE " "label from the options below and report your confidence that this label is correct.\n" "Respond with ONLY a single-line JSON object in this exact format, no other text, no " 'reasoning, no markdown fences:\n{"answer": "<LETTER>", "confidence": <number between 0 ' "and 1>}\n\n" "Input: add this song to my road trip playlist\n\n" "Options:\nA) BookRestaurant\nB) AddToPlaylist\nC) GetWeather\nD) PlayMusic\n" "E) RateBook\nF) SearchCreativeWork\nG) SearchScreeningEvent" )}], tokenize=False, add_generation_prompt=True, enable_thinking=False,)