Qwen AgentWorld 35B A3B, GGUF
Qwen AgentWorld 35B A3B is a native language world model designed to simulate agent environments across MCP, Search, Terminal, SWE, Android, Web, and OS interaction domains. It features a hybrid architecture. The model combines Gated DeltaNet and Gated Attention. Trained through CPT, SFT, and RL, it supports native world modeling.
This repository holds Vultr's GGUF quantizations of it. They are hybrid quantizations: rather than applying one precision to the whole network, each layer's precision is chosen from a measured importance score, so the bits go where they change the output most.
Model Overview
| Attribute | Value |
|---|---|
| Base model | Qwen/Qwen-AgentWorld-35B-A3B |
| Architecture | qwen35moe, Mixture-of-Experts |
| Parameters | 35B total |
| Active parameters | 3B |
| Layers | 40 |
| Hidden size | 2,048 |
| Attention heads | 16 |
| Experts | 256 routed, 8 selected per token, shared expert |
| Context length | 262K tokens |
| Vocabulary | 248,320 |
| Attention layout | 3:1 ratio of Gated DeltaNet to Gated Attention |
| License | Apache 2.0 |
| Attribute | Value |
|---|---|
| Base model | Qwen/Qwen-AgentWorld-35B-A3B |
| Architecture | qwen35moe, Mixture-of-Experts |
| Parameters | 35B total |
| Active parameters | 3B |
| Layers | 40 |
| Hidden size | 2,048 |
| Attention heads | 16 |
| Experts | 256 routed, 8 selected per token, shared expert |
| Context length | 262K tokens |
| Vocabulary | 248,320 |
| Attention layout | 3:1 ratio of Gated DeltaNet to Gated Attention |
| License | Apache 2.0 |
Quantizations
| File | Type | Size | KLD vs bf16 |
|---|---|---|---|
| Qwen-AgentWorld-35B-A3B-Q5_K_M.gguf | Q5_K_M | 27.19 GB | 0.0075 |
| Qwen-AgentWorld-35B-A3B-Q4_K_M.gguf | Q4_K_M | 23.26 GB | 0.0151 |
| Qwen-AgentWorld-35B-A3B-IQ3_M.gguf | IQ3_M | 16.97 GB | 0.0406 |
| Qwen-AgentWorld-35B-A3B-IQ3_S.gguf | IQ3_S | 16.77 GB | 0.0449 |
| File | Type | Size | KLD vs bf16 |
|---|---|---|---|
| Qwen-AgentWorld-35B-A3B-Q5_K_M.gguf | Q5_K_M | 27.19 GB | 0.0075 |
| Qwen-AgentWorld-35B-A3B-Q4_K_M.gguf | Q4_K_M | 23.26 GB | 0.0151 |
| Qwen-AgentWorld-35B-A3B-IQ3_M.gguf | IQ3_M | 16.97 GB | 0.0406 |
| Qwen-AgentWorld-35B-A3B-IQ3_S.gguf | IQ3_S | 16.77 GB | 0.0449 |
KL divergence measures how far a quantized model's next-token probabilities drift from those of the original bf16 model, averaged over every token of the evaluation text. The lower the value, the closer the quantized model is to the original; zero would mean identical output.
Evaluation
Every variant in this repository was evaluated against the original bf16 model for perplexity, KL divergence, top-token agreement, HellaSwag and WinoGrande, under identical conditions. The full results are maintained on the model's page in the Vultr Model Library:
https://docs.vultr.com/models/alibaba-qwen-agentworld/qwen-agentworld-35b-a3b?tab=quantized
That page is the reference for these files. The summary above is an excerpt from it.
Run
llama-server -hf vultr/Qwen-AgentWorld-35B-A3B-GGUF:Q4_K_M
The model's chat template is embedded in every file, so no extra prompt formatting is needed.
License
These files are derived from Qwen/Qwen-AgentWorld-35B-A3B and are distributed under the same license, Apache 2.0. The model is the work of Qwen; Vultr produced the quantizations.