GLM-5.2 is live. #1 throughput on OpenRouter, pay-per-token on FriendliAI. Try it today ➜
Matmultoken
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
Matmultoken
Model Tree
Base
Qwen/Qwen3.5-4B
Quantized
this model
Input Modalities
Text
Image
Video
Output Modalities
Text
Supported Functionality
Dedicated Endpoints
| field | value |
|---|---|
| base model | Qwen/Qwen3.5-4B |
| modality | text |
| common_dim | 2560 |
| rank | 32 |
| mine_layers | 16 (overhead dial; layer count) |
| pipeline | vllm |
Text LLMs mine during prefill — when many tokens are processed at once (rows = tokens is large). Single-token decode does not mine (rows ≈ 1), so interactive chat mines far less than long-prompt or batched-prefill serving. Diffusion models mine on every forward (large token count always), so for continuous mining a diffusion model (see Matmultoken/Z-Image-Turbo-pouw) is the stronger substrate; this LLM repo is for prefill-heavy / batch workloads.
python
# Serve via vLLM with quantization="pouw" (vLLM-MatMulToken plugin auto-registers it).from vllm import LLMllm = LLM(model="Matmultoken/Qwen3.5-4B-pouw", quantization="pouw") # mines on eligible matmuls while it servesprint(llm.generate("The history of money is")) # generation is bit-identical to the base model
Matmultoken organization. The base weights (apache-2.0) are bundled in this repo at a pinned snapshot for a reproducible mining shape; the original model's LICENSE and attribution are preserved in-repo.Generated by MatMulToken publish_pouw_models.py. License: MIT.