GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
Hyukkyu
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
Hyukkyu
Model Tree
Qwen/Qwen3-8B-Base
Input Modalities
Output Modalities
Supported Functionality
| Split | Correct / total | Accuracy |
|---|---|---|
| Native forget | 399/400 | 99.75% |
| Native retain | 400/400 | 100.00% |
| Paraphrased forget | 255/400 | 63.75% |
| Paraphrased retain | 254/400 | 63.50% |
| RAQUEL affected | 529/1316 | 40.20% |
| RAQUEL unaffected | 1114/2518 | 44.24% |
All table rows were rejudged with pinned Qwen/Qwen3.8-27B at temperature 0 with thinking disabled. Native and paraphrased cohorts contain 400 queries each. RAQUEL accuracy covers every affected query (1,316) and every unaffected query (2,518), with no sampling. The questions were excluded from configuration tuning. Earlier judge results remain in evaluation.json for provenance.
Qwen/Qwen3-8B-Base, revision 49e3418fbbbca6ecbdf9608b4d22e5a407081db4.Hyukkyu/RAQUEL2-ICLR, revision aee9a541f1ba58032e2eb62a1e56149475cd1126; native TOFU uses the frozen campaign inputs derived from locuslab/TOFU.Question: {question}\nAnswer:; answer-only QA loss; masked right padding trimmed without removing supervised tokens.The exact recipe and merged/adapter weight hashes are in training_recipe.json. package_versions.json records the local runtime used to prepare this release. The adapter config uses the public base repository and its pinned revision rather than a machine-local path.
python
import torchfrom transformers import AutoModelForCausalLM, AutoTokenizer repo_id = "Hyukkyu/Qwen3-8B-Base-RAQUEL-TOFU-M-orig-LoRA-v1"tokenizer = AutoTokenizer.from_pretrained(repo_id)model = AutoModelForCausalLM.from_pretrained( repo_id, dtype=torch.bfloat16, device_map="auto").eval()prompt = "Question: What is the author's full name?\nAnswer:"inputs = tokenizer(prompt, return_tensors="pt").to(model.device)with torch.inference_mode(): output = model.generate( **inputs, do_sample=False, max_new_tokens=96, pad_token_id=tokenizer.eos_token_id, )print(tokenizer.decode(output[0, inputs.input_ids.shape[1]:], skip_special_tokens=True))Use the plain QA prefix above for benchmark evaluation. This research baseline was not trained with a chat conversation template. Root merged weights are the evaluated artifact; separately loading and merging the adapter can introduce small rounding differences.
The adapter may be loaded with PeftModel.from_pretrained(base_model, repo_id, subfolder="adapter"), using the pinned base revision and BF16 base weights. Only model/tokenizer files, the portable adapter, aggregate evaluation, recipe, package versions, and upstream notices are included.
import torchfrom transformers import AutoModelForCausalLM, AutoTokenizer repo_id = "Hyukkyu/Qwen3-8B-Base-RAQUEL-TOFU-M-orig-LoRA-v1"tokenizer = AutoTokenizer.from_pretrained(repo_id)model = AutoModelForCausalLM.from_pretrained( repo_id, dtype=torch.bfloat16, device_map="auto").eval()prompt = "Question: What is the author's full name?\nAnswer:"inputs = tokenizer(prompt, return_tensors="pt").to(model.device)with torch.inference_mode(): output = model.generate( **inputs, do_sample=False, max_new_tokens=96, pad_token_id=tokenizer.eos_token_id, )print(tokenizer.decode(output[0, inputs.input_ids.shape[1]:], skip_special_tokens=True))