GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
buugiiiiii
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
buugiiiiii
Model Tree
zai-org/GLM-4.7-Flash
Input Modalities
Output Modalities
Supported Functionality
| benchmark | base | K-Guard 12.5% | K-Guard 25% | K-Guard 50% | REAP-Code 25% |
|---|---|---|---|---|---|
| Korean (held-out harness) | |||||
| KMMLU | 49.4 | 48.8 | 42.4 | 23.6 | 35.0 |
| KoBEST BoolQ | 90.2 | 87.7 | 78.8 | 74.3 | 79.6 |
| KoBEST COPA | 70.7 | 69.7 | 65.0 | 55.9 | 51.7 |
| KoBEST SentiNeg | 74.8 | 71.8 | 70.5 | 53.7 | 49.6 |
| KoBEST HellaSwag | 58.8 | 55.0 | 57.4 | 48.8 | 51.6 |
| English reasoning / knowledge | |||||
| ARC-Challenge | 87.1 | 86.1 | 79.9 | 36.9 | 79.1 |
| HellaSwag (en) | 74.2 | 67.4 | 55.3 | 32.6 | 61.7 |
| WinoGrande | 67.4 | 65.6 | 59.2 | 53.1 | 63.5 |
| MMLU-Pro | 41.2 | 36.9 | 34.2 | 14.8 | 33.8 |
| TruthfulQA MC2 | 47.3 | 46.0 | 43.1 | 40.6 | 44.6 |
| MMLU-en | 69.7 | 64.5 | 56.6 | 30.1 | 59.6 |
| Math | |||||
| GSM8K | 86.3 | 86.7 | 81.8 | 61.3 | 86.7 |
| MathQA | 30.5 | 29.1 | 24.8 | 21.3 | 28.7 |
| Coding | |||||
| HumanEval+ | 81.1 | 75.6 | 74.4 | 65.2 | 80.5 |
| MBPP+ | 70.9 | 69.3 | 67.5 | 61.6 | 67.5 |
| Safety / bias | |||||
| BBQ | 76.6 | 69.9 | 59.8 | 33.6 | 74.0 |
| SafetyBench | 74.8 | 73.0 | 69.1 | 44.3 | 72.1 |
| Macro | |||||
| Korean MC macro | 68.8 | 66.6 | 62.8 | 51.2 | 53.5 |
| Academic 10-bench macro | 65.5 | 62.5 | 56.4 | 36.9 | 60.4 |
| Coding 2-bench macro | 76.0 | 72.5 | 70.9 | 63.4 | 74.0 |
Glm4MoeLiteForCausalLM.modeling_*.py is
required, and the checkpoint loads with trust_remote_code=False.topk_method=noaux_tc,
norm_topk_prob=true, router correction bias, and routed_scaling_factor=1.8.mtp_compatible=false. Use standard autoregressive generation; speculative
MTP/NextN decoding is not supported by this release.7dd20894a642a0aa287e9827cb1a1f7f91386b67| Source | Records | Observed tokens | Role |
|---|---|---|---|
theblackcat102/evol-codealpaca-v1 | 8,192 | 4,366,438 | Code-centric standard REAP baseline |
[experts, tokens, hidden] activation tensor was retained.64 → 48.These are decision-grade experimental results, not release-certified leaderboard numbers:
decision_eligible=true, release_eligible=false; do not compare these subset scores directly
with full-set or few-shot leaderboard values.Machine-readable results and the exact survivor manifest are included under evaluation/.
Certification status: This is a controlled code-centric REAP baseline. K-Guard certification criteria do not apply because Korean hard protection was intentionally omitted.
Install a Transformers release that includes Glm4MoeLiteForCausalLM:
bash
pip install "transformers>=5.0.0" accelerate safetensorsThe checkpoint was validated with Transformers 5.0.0 and NVIDIA PyTorch 25.09. It uses the native Transformers model implementation and does not execute custom code from this repository.
python
import torchfrom transformers import AutoModelForCausalLM, AutoTokenizer model_id = "buugiiiiii/GLM-4.7-Flash-REAP-Code-48E"tokenizer = AutoTokenizer.from_pretrained( model_id, trust_remote_code=False,)model = AutoModelForCausalLM.from_pretrained( model_id, dtype=torch.bfloat16, device_map="auto", trust_remote_code=False, attn_implementation="eager", experts_implementation="eager",)model.eval() messages = [{"role": "user", "content": "한국어로 이진 탐색을 설명해줘."}]prompt = tokenizer.apply_chat_template( messages, tokenize=False, add_generation_prompt=True, enable_thinking=False,)inputs = tokenizer(prompt, return_tensors="pt").to(model.device) with torch.inference_mode(): output = model.generate( **inputs, max_new_tokens=256, do_sample=False, pad_token_id=tokenizer.eos_token_id, ) print(tokenizer.decode(output[0, inputs["input_ids"].shape[1]:], skip_special_tokens=True))Expert pruning is irreversible within this checkpoint and can change routing behavior sharply on tasks not represented by the calibration or evaluation sets. No post-pruning recovery or fine-tuning was applied. Reported Korean behavior is relative to the tested base checkpoint and harness; it does not imply universal Korean-language safety or quality preservation. The current release is not validated for speculative MTP/NextN decoding.
evaluation/glm-full-b19-results.json: machine-readable five-model benchmark resultsevaluation/glm-full-b19-long-scores.csv: long-format benchmark scoresevaluation/survivor-manifest.json: exact per-layer old Expert IDs retainedartifacts/calibration-validation.json: independent calibration-manifest validationartifacts/observation-validation.json: selected-only observation validationartifacts/expert-statistics.csv: per-Expert scoring/features used for this method7dd20894a642a0aa287e9827cb1a1f7f91386b677f7ac2f767e68c61528082e62c4eef3a6027429ab1e47b654de10b4c93b760d6PASS and zero validation failures.Please cite the original GLM work and the K-Guard/K-REAP repository when using this checkpoint.
import torchfrom transformers import AutoModelForCausalLM, AutoTokenizer model_id = "buugiiiiii/GLM-4.7-Flash-REAP-Code-48E"tokenizer = AutoTokenizer.from_pretrained( model_id, trust_remote_code=False,)model = AutoModelForCausalLM.from_pretrained( model_id, dtype=torch.bfloat16, device_map="auto", trust_remote_code=False, attn_implementation="eager", experts_implementation="eager",)model.eval() messages = [{"role": "user", "content": "한국어로 이진 탐색을 설명해줘."}]prompt = tokenizer.apply_chat_template( messages, tokenize=False, add_generation_prompt=True, enable_thinking=False,)inputs = tokenizer(prompt, return_tensors="pt").to(model.device) with torch.inference_mode(): output = model.generate( **inputs, max_new_tokens=256, do_sample=False, pad_token_id=tokenizer.eos_token_id, ) print(tokenizer.decode(output[0, inputs["input_ids"].shape[1]:], skip_special_tokens=True))