GLM-5.2 is live. #1 throughput on OpenRouter, pay-per-token on FriendliAI. Try it today ➜
Icey444
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Run this model inference with full control and performance in your environment.
Model Details
Model Provider
Icey444
Model Tree
Qwen/Qwen3-Embedding-4B
Input Modalities
Output Modalities
Supported Functionality
LoRA adapter for Qwen/Qwen3-Embedding-4B — Relevance-SFT (direct; winner recipe).
Trained on multi-hop MuSiQue; evaluated on MuSiQue (in-domain) + 2WikiMultiHopQA, HotpotQA, NQ, TriviaQA (OOD).
Indirect feedback = the reward is a frozen reader LLM's answer success (log p(gold answer | query, retrieved context)) — no passage-level relevance labels. Rel-SFT = the direct-supervision baseline (contrastive InfoNCE on gold passages).
Full results (retrieval R@k + downstream RAG accuracy) and training scripts: https://irisicy4.github.io/exp-record/projects/indirect-rag/
python
from peft import PeftModelfrom transformers import AutoModel, AutoTokenizerbase = AutoModel.from_pretrained("Qwen/Qwen3-Embedding-4B")model = PeftModel.from_pretrained(base, "Icey444/emb-rag-qwen4b-relsft").merge_and_unload()tok = AutoTokenizer.from_pretrained("Icey444/emb-rag-qwen4b-relsft")