GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
Akicou
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
Akicou
Model Tree
zai-org/GLM-4.7-Flash
Input Modalities
Output Modalities
Supported Functionality
Important Update regarding GGUF support:
A critical bug was recently identified in llama.cpp regarding the scoring_func for GLM models (which caused looping and poor output quality). While the base weights are functional, the GGUF files for this model are currently being re-generated to ensure full compatibility with the latest fixes.
llama.cpp or Unsloth, please refer to the official Unsloth GLM-4.7-Flash documentation for the most stable configuration parameters.transformers and vLLM for immediate use.50% Expert-Pruned GLM-4.7 Flash optimized for code generation, function calling, and agentic workflows.
Created using REAP (Router-weighted Expert Activation Pruning) by Cerebras:
--
text
REAP Algorithm:1. Forward pass calibration samples through model2. Record which experts activate and their magnitudes3. Compute saliency = router_weight × activation_norm4. Prune lowest-saliency experts Key Insight: Experts are TASK-SPECIFIC├── Some experts specialize in natural language├── Some experts specialize in code syntax├── Some experts specialize in JSON/structured output└── Some experts specialize in multi-turn context If calibration lacks code → code-specialized experts appear "unused" → get pruned → model loses coding abilityCerebras used the same 3 datasets in their GLM-4.6 REAP experiments:
We followed this exact recipe for reproducibility.
bash
vllm serve Akicou/GLM-4.7-Flash-REAP-19 \ --tensor-parallel-size 2 \ --trust-remote-code \ --dtype bfloat16python
import torchfrom transformers import AutoModelForCausalLM, AutoTokenizer model = AutoModelForCausalLM.from_pretrained( "Akicou/GLM-4.7-Flash-REAP-19", torch_dtype=torch.bfloat16, device_map="auto", trust_remote_code=True)tokenizer = AutoTokenizer.from_pretrained("Akicou/GLM-4.7-REAP-19", trust_remote_code=True) messages = [{"role": "user", "content": "Write a Python function to merge two sorted lists."}]inputs = tokenizer.apply_chat_template(messages, return_tensors="pt", add_generation_prompt=True)outputs = model.generate(inputs.to(model.device), max_new_tokens=512, temperature=0.7)print(tokenizer.decode(outputs[0], skip_special_tokens=True))MIT (inherited from GLM-4.7 Flash)
bibtex
@article{lasby2025reap, title={REAP the Experts: Why Pruning Prevails for One-Shot MoE Compression}, author={Lasby, Mike and Lazarevich, Ivan and Sinnadurai, Nish and Lie, Sean and Ioannou, Yani and Thangarasa, Vithursan}, journal={arXiv preprint arXiv:2510.13999}, year={2025}, url={https://arxiv.org/abs/2510.13999}}REAP Algorithm:1. Forward pass calibration samples through model2. Record which experts activate and their magnitudes3. Compute saliency = router_weight × activation_norm4. Prune lowest-saliency expertsKey Insight: Experts are TASK-SPECIFIC├── Some experts specialize in natural language├── Some experts specialize in code syntax├── Some experts specialize in JSON/structured output└── Some experts specialize in multi-turn contextIf calibration lacks code → code-specialized experts appear "unused" → get pruned → model loses coding abilityvllm serve Akicou/GLM-4.7-Flash-REAP-19 \ --tensor-parallel-size 2 \ --trust-remote-code \ --dtype bfloat16import torchfrom transformers import AutoModelForCausalLM, AutoTokenizer model = AutoModelForCausalLM.from_pretrained( "Akicou/GLM-4.7-Flash-REAP-19", torch_dtype=torch.bfloat16, device_map="auto", trust_remote_code=True)tokenizer = AutoTokenizer.from_pretrained("Akicou/GLM-4.7-REAP-19", trust_remote_code=True) messages = [{"role": "user", "content": "Write a Python function to merge two sorted lists."}]inputs = tokenizer.apply_chat_template(messages, return_tensors="pt", add_generation_prompt=True)outputs = model.generate(inputs.to(model.device), max_new_tokens=512, temperature=0.7)print(tokenizer.decode(outputs[0], skip_special_tokens=True))@article{lasby2025reap, title={REAP the Experts: Why Pruning Prevails for One-Shot MoE Compression}, author={Lasby, Mike and Lazarevich, Ivan and Sinnadurai, Nish and Lie, Sean and Ioannou, Yani and Thangarasa, Vithursan}, journal={arXiv preprint arXiv:2510.13999}, year={2025}, url={https://arxiv.org/abs/2510.13999}}