GLM-5.2 is live. #1 throughput on OpenRouter, pay-per-token on FriendliAI. Try it today ➜
AdarshSingh7647
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
AdarshSingh7647
Model Tree
Qwen/Qwen3-8B
Input Modalities
Output Modalities
Supported Functionality
System prompt used at eval time:
You are a table relevance expert. Given a question and a set of candidate tables, rank them from most to least useful for answering the question. Reason in a
<think>...</think>block, then output exactly JSON with keyranked_tables.
The user message lists candidate tables as ### Table 1, ### Table 2, … The
model outputs, after a <think> reasoning block, a JSON object:
json
{"ranked_tables": [3, 1, 5, 2, 4, ...]}
where the integers are the 1-based candidate positions in the prompt, best first. Map them back to your table ids to obtain the reranked list.
Evaluated as a single-call listwise reranker (think mode) reordering the first-stage top-25 candidate list, on 4 out-of-domain table-QA datasets. SQA / TAT-QA use the full test set; HybridQA / TabFact use a fixed shared 500-query sample.
| Dataset | n | recall@10 | ndcg@10 | acc@10 | MRR |
|---|---|---|---|---|---|
| SQA | 148 | 0.893 | 0.764 | 0.838 | 0.735 |
| TAT-QA | 362 | 0.666 | 0.546 | 0.431 | 0.637 |
| HybridQA | 500 | 0.882 | 0.790 | 0.796 | 0.833 |
| TabFact | 500 | 0.809 | 0.726 | 0.634 |
acc@k = all gold tables within the top-k (strict).
python
from transformers import AutoModelForCausalLM, AutoTokenizermodel_id = "CoTCond_merged" # or your uploaded repo idtok = AutoTokenizer.from_pretrained(model_id)model = AutoModelForCausalLM.from_pretrained(model_id, torch_dtype="bfloat16", device_map="auto")SYSTEM = ("You are a table relevance expert. Given a question and a set of candidate ""tables, rank them from most to least useful for answering the question. ""Reason in a <think>...</think> block, then output exactly JSON with key ranked_tables.")user = "Question: ...\n\n### Table 1\n...\n\n### Table 2\n...\n" # your listwise promptmsgs = [{"role": "system", "content": SYSTEM}, {"role": "user", "content": user}]text = tok.apply_chat_template(msgs, tokenize=False, add_generation_prompt=True)out = model.generate(**tok(text, return_tensors="pt").to(model.device),max_new_tokens=8192, temperature=0.6, top_p=0.95)print(tok.decode(out[0], skip_special_tokens=True))
For high-throughput reranking use vLLM (loads directly, no LoRA flags).
model.safetensors (~16 GB).The accompanying paper is currently under review; a citation will be added here once it is available on arXiv.
| 0.817 |
| Mean | 0.813 | 0.706 | 0.675 | 0.756 |