AlibabaQwenQwen3_5MoeForConditionalGeneration

Qwen AgentWorld 35B A3B — Hardware Requirements & GPU Compatibility

ChatFunctions

Qwen AgentWorld 35B A3B is Alibaba's 35-billion-parameter mixture-of-experts model, with about 3 billion parameters active per token (the A3B in its name). Unlike a typical chat model, it is a language world model: given an agent's action and history, it predicts what the environment does next, covering domains such as terminal, web, Android, and software engineering. Only the active experts run per token, so inference stays fast even though all the weights must fit in memory. At this size, local inference calls for quantization and a single high-end consumer GPU. The model supports a 262K token context window and is released under the Apache 2.0 license, allowing unrestricted commercial and research use. Published in June 2026, it is built on a Qwen3.5-35B-A3B base and is meant for simulating agents, not direct conversation.

34.7B params · MoE3.5B active262K contextApache 2.0Jun 20267.4K downloads739 likes297.5K quant downloads

Based on Qwen3.5 35B A3B Base

Your hardware

Verdicts below are for the reference RTX 4090. Set yours to personalise the page.

Detection reads your browser's GPU name with no data sent anywhere.

Runs great on a 24 GB GPU

Runs from 12 GB (IQ2_XXS · 128K)Best on 24 GB (Q4_K_M · ~184 tok/s · 256K)Macs from 16 GB (IQ2_M · ~60 tok/s)

Apache 2.0

Reference GPU · RTX 4090 24 GB

Estimates, ±20% · how we estimate

Run Qwen AgentWorld 35B A3B

One recommended way. Every other method, OS and quant is in the step-by-step guide.

Other ways

each opens step-by-step

Run in cloud

Fits on RTX A6000 (48 GB) (26 GB headroom) · Q4_K_M

Generation speed: ~156 tok/s; generation speed
Cost per 1M output tokens: $0.59; per 1M output tokens

Deployon RunPodBest optionfrom $0.33/hr

Compare GPUs →

Run Qwen AgentWorld 35B A3B

21.3 GB VRAM at 16K context · fits the RTX 4090 (2.7 GB free)

Works on macOS, Windows and Linux.

Install Ollama (Linux)
$curl -fsSL https://ollama.com/install.sh | sh
Install Ollama (macOS / Windows)
Download and run the app; it installs the ollama command.
Download Ollama ↗
Run Q4_K_M
$ollama run hf.co/unsloth/Qwen-AgentWorld-35B-A3B-GGUF:Q4_K_M
Raise the context window
Type it inside the chat. For the API server instead, start it with OLLAMA_CONTEXT_LENGTH=16384 ollama serve.
$/set parameter num_ctx 16384
A larger context needs more memory than the weights alone.
Use the API
Ollama serves its API on http://localhost:11434.

Works on macOS, Windows and Linux.

Install LM Studio
Desktop app for macOS (Apple silicon), Windows and Linux.
lmstudio.ai ↗
Download the model
In the app press Ctrl/Cmd + Shift + M and paste https://huggingface.co/unsloth/Qwen-AgentWorld-35B-A3B-GGUF, then pick Q4_K_M. Or use the CLI (run LM Studio once first):
$lms get https://huggingface.co/unsloth/Qwen-AgentWorld-35B-A3B-GGUF@Q4_K_M
Open on Hugging Face ↗
Load and chat
Open the model loader, select the model and set Context Length to 16384. Leave GPU offload at maximum if it fits.
Start the local server (optional)
$lms server start

Works on macOS, Windows and Linux.

Install llama.cpp (macOS)
MacPorts, Nix and conda-forge also work.
$brew install llama.cpp
Install options ↗
Install llama.cpp (Windows)
$winget install llama.cpp
Install llama.cpp (Linux)
Download the prebuilt llama-*-bin-ubuntu-* archive for your GPU backend (CPU, Vulkan) from the releases page, unpack it and run the binaries from that folder. Homebrew (brew install llama.cpp) and conda-forge also work on Linux.
llama.cpp releases ↗
Start the server
Downloads the model on first run. The chat UI and an OpenAI-compatible API are at http://localhost:8080.
$llama-server -hf unsloth/Qwen-AgentWorld-35B-A3B-GGUF:Q4_K_M -c 16384 -ngl 99 --port 8080

Works on macOS, Windows and Linux.

Option A: Docker Model Runner
Docker Desktop: Settings > AI > Enable Docker Model Runner. Docker Engine on Linux: install the docker-model-plugin package.
$docker model run hf.co/unsloth/Qwen-AgentWorld-35B-A3B-GGUF:Q4_K_M
Model Runner docs ↗
The quant tag after the repo is how Hugging Face cards write it; if your Docker version rejects it, drop the tag and pick the quant by repo.
Option B: llama.cpp server image (NVIDIA)
Needs the NVIDIA Container Toolkit. Models are cached in the llama-cache volume. Open http://localhost:8080.
$docker run --gpus all -p 8080:8080 -v llama-cache:/root/.cache ghcr.io/ggml-org/llama.cpp:server-cuda -hf unsloth/Qwen-AgentWorld-35B-A3B-GGUF:Q4_K_M -c 16384 -ngl 99 --host 0.0.0.0 --port 8080
Image tags (cuda13, vulkan, rocm) ↗
Cache path inside the image is not in the docs; remove the -v flag if the volume complains.

Linux only.

Install vLLM
Linux only; on Windows run it inside WSL2. NVIDIA GPU shown; AMD ROCm, Intel and CPU builds have their own install pages.
$uv venv --python 3.12 --seed
$source .venv/bin/activate
$uv pip install vllm --torch-backend=auto
vLLM installation ↗
Serve the model
OpenAI-compatible API at http://localhost:8000/v1.
$vllm serve Qwen/Qwen-AgentWorld-35B-A3B --max-model-len 16384
Runs the original weights, not the Q4_K_M GGUF; needs ~76 GB of GPU memory.

macOS only.

MLX supports only macOS, so follow these steps for that setup.

Install mlx-lm
$pip install mlx-lm
Chat in the terminal
$mlx_lm.chat --model Brooooooklyn/Qwen-AgentWorld-35B-A3B-nvfp4-mlx
Downloads the model on first run. The MLX repo has its own quantisation; it may not match the quant selected above.
Or start a local server
Listens on http://localhost:8080.
$mlx_lm.server --model Brooooooklyn/Qwen-AgentWorld-35B-A3B-nvfp4-mlx

Works on macOS, Windows and Linux.

Install Jan
jan.ai ↗
Find the model
Open the Hub tab, paste this model ID in Search, and choose Q4_K_M if offered.
$unsloth/Qwen-AgentWorld-35B-A3B-GGUF
Use the local API (optional)
Settings > Local API Server > Start Server. It listens on http://127.0.0.1:1337.

Rent GPUs hourly to run Q4_K_M right there. Pricing comes from partner marketplaces and varies.

Available GPU rentals for Q4_K_M, prioritized by value
GPU	Speed	$/1M tokens	Rent
RTX A6000 (48 GB)CheapestCheapest26 GB free	~156 tok/s	$0.59	Deploy on RunPodfrom $0.33/hr (opens in new tab)	48 GB · 26 GB free · ~156 tok/s · $0.59 / 1M tokens
RTX PRO 4500 (32 GB)10 GB free	~172 tok/s	$0.55	Deploy on RunPodfrom $0.34/hr (opens in new tab)	32 GB · 10 GB free · ~172 tok/s · $0.55 / 1M tokens
A40 (48 GB)26 GB free	~146 tok/s	$0.67	Deploy on RunPodfrom $0.35/hr (opens in new tab)	48 GB · 26 GB free · ~146 tok/s · $0.67 / 1M tokens
RTX A6000 (48 GB)26 GB free	~156 tok/s	$0.72	Deploy on Vast.aifrom $0.40/hr (opens in new tab)	48 GB · 26 GB free · ~156 tok/s · $0.72 / 1M tokens
RTX 5000 Ada (32 GB)10 GB free	~128 tok/s	$1.07	Deploy on RunPodfrom $0.49/hr (opens in new tab)	32 GB · 10 GB free · ~128 tok/s · $1.07 / 1M tokens
A40 (48 GB)26 GB free	~146 tok/s	$1.02	Deploy on Vast.aifrom $0.54/hr (opens in new tab)	48 GB · 26 GB free · ~146 tok/s · $1.02 / 1M tokens
RTX 6000 Ada (48 GB)26 GB free	~179 tok/s	$1.00	Deploy on Vast.aifrom $0.64/hr (opens in new tab)	48 GB · 26 GB free · ~179 tok/s · $1.00 / 1M tokens
A100 40GB18 GB free	~233 tok/s	$0.88	Deploy on Vast.aifrom $0.73/hr (opens in new tab)	18 GB free · ~233 tok/s · $0.88 / 1M tokens

Compare GPUs →

How Much VRAM Does Qwen AgentWorld 35B A3B Need?

Select a quantization level; the listed hardware adheres to it.

VRAM	+32K	+128K	File	Quality	On RTX 4090 (reference)
9.9 GB	10.2 GB	11.2 GB	11.5 GB	2-bit · i-quant extreme compression — Importance-weighted 2-bit, extreme compression — significant quality loss	Fits: 256K context · ~270 tok/s
12.1 GB	12.3 GB	13.3 GB	11.6 GB	2-bit · i-quant medium — Importance-weighted 2-bit, medium	Fits: 256K context · ~248 tok/s
13.8 GB	14.1 GB	15.1 GB	13.7 GB	3-bit · i-quant — Importance-weighted 3-bit	Fits: 256K context · ~233 tok/s
15.1 GB	15.4 GB	16.4 GB	15 GB	3-bit · i-quant small — Importance-weighted 3-bit, small	Fits: 256K context · ~222 tok/s
15.1 GB	15.4 GB	16.4 GB	12.9 GB	2-bit · K-quant — 2-bit quantization with K-quant improvements	Fits: 256K context · ~222 tok/s
15.6 GB	15.8 GB	16.8 GB	15.2 GB	3-bit small — 3-bit small quantization	Fits: 256K context · ~219 tok/s
17.3 GB	17.5 GB	18.5 GB	16.7 GB	3-bit medium — 3-bit medium quantization	Fits: 256K context · ~207 tok/s
18.1 GB	18.4 GB	19.4 GB	18.1 GB	3-bit large — 3-bit large quantization	Fits: 256K context · ~202 tok/s
19.0 GB	19.3 GB	20.3 GB	17.8 GB	4-bit · i-quant compact — Importance-weighted 4-bit, compact	Fits: 256K context · ~196 tok/s
19.9 GB	20.1 GB	21.1 GB	20.9 GB	4-bit small — 4-bit small quantization	Fits: 256K context · ~191 tok/s
19.9 GB	20.1 GB	21.1 GB	18.1 GB	4-bit · i-quant non-linear — Importance-weighted 4-bit, non-linear	Fits: 256K context · ~191 tok/s
21.2 GB	21.4 GB	22.4 GB	22.1 GB	4-bit medium — 4-bit medium quantization — most popular sweet spot	Fits: 256K context · ~184 tok/s
24.2 GB	24.5 GB	25.5 GB	24.9 GB	5-bit small — 5-bit small quantization	Does not fit: needs 24.3 GB
25.1 GB	25.3 GB	26.3 GB	26.5 GB	5-bit medium — 5-bit medium quantization — good quality/size tradeoff	Does not fit: needs 25.2 GB
29.0 GB	29.2 GB	30.2 GB	29.3 GB	6-bit, very good — 6-bit quantization, very good quality	Does not fit: needs 29.1 GB
35.0 GB	35.3 GB	36.3 GB	36.9 GB	8-bit, near-lossless — 8-bit quantization, near-lossless	Does not fit: needs 35.1 GB
69.7 GB	70 GB	71 GB	69.4 GB	16-bit · bfloat — Brain floating point 16 — preferred for training	Does not fit: needs 69.8 GB

Selected Q4_K_M

4-bit medium quantization

21.3GBat 16K context

Weights: 20.8 GB
KV cache: 0.2 GB
Overhead: 0.3 GB

Fits RTX 4090 (reference) · 2.7 GB free · ~184 tok/s

≥50 tok/s needs ≥182 GB/s · RTX 5090 ~249, RTX 3090 Ti ~184

Which Hardware Can Run Qwen AgentWorld 35B A3B?

Q4_K_M · 21.2 GB

Qwen AgentWorld 35B A3B (Q4_K_M) requires 21.2 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 28+ GB is recommended. Using the full 262K context window can add up to 2.6 GB, bringing total usage to 23.8 GB. 7 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti. 100 devices with unified memory can run Qwen AgentWorld 35B A3B, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB), Mac Studio M1 Max (32 GB).

By VRAM tier · typical GPUs and Macs at Q4_K_M

Typical hardwareQuantVRAMtok/sContext that fits

8 GB · RTX 4060Doesn't fit — smallest quant needs 10 GB8 GB tier →

12 GB · RTX 3060 12GBIQ2_XXS · 10 GB · ~156 tok/s · 128K10~156128K12 GB tier →

16 GB · RTX 5060 Ti 16GBQ3_K_S · 15.6 GB · ~133 tok/s · 32K15.6~13332K16 GB tier →

24 GB · RTX 4090referenceQ4_K_M · 21.3 GB · ~184 tok/s · 256K21.3~184256K24 GB tier →

Mac 32 GB · Mac Mini M4Q4_K_M · 21.3 GB · ~30 tok/s · 256K21.3~30256KMacs →

Mac 64 GB · Mac Studio M4 MaxQ4_K_M · 21.3 GB · ~74 tok/s · 256K21.3~74256KMacs →

Alternatives to Qwen AgentWorld 35B A3B

Qwen family · 28 models →

Similar models

6tok/s on RTX 4090

Model	Score	VRAM	tok/s	Why it is similar
MiniCPM V 4.7 35B A3BOpenBMB	—	22.1 GB	~169	same size
Llm Jp 4.1 32B A3b Thinkingllm-jp	—	20.1 GB	~188	same size
Llm Jp 4.1 33B Thinkingllm-jp	—	22.5 GB	~35	same size
Agnes 3.0 FlashAgnes-AI	—	22.1 GB	~35	same size
Ornith 1.5 35B A3Bornith-ai	—	22 GB	~151	same size
Agents A1InternScience	—	21.5 GB	~171	same size

Fine-tunes & variants(4)community builds of Qwen AgentWorld 35B A3B

Specifications

Publisher: Alibaba
Family: Qwen
Parameters: 34.7B
Active parameters: 3.5B per token
Architecture: Qwen3_5MoeForConditionalGeneration
Context Length: 262,144 tokens
Vocabulary Size: 248,320
Release Date: 2026-06-22
License: Apache 2.0

Resources

Frequently Asked Questions

How much VRAM does Qwen AgentWorld 35B A3B need?

Qwen AgentWorld 35B A3B requires 21.2 GB of VRAM at Q4_K_M, or 69.7 GB at BF16. Full 262K context adds up to 2.6 GB (23.8 GB total).

VRAM = Weights + KV Cache + Overhead

Weights = 34.7B × 4.8 bits ÷ 8 = 20.8 GB

KV Cache + Overhead ≈ 0.4 GB (at 2K context + ~0.3 GB framework)

Fit ratings and hardware model lists check this model with room for a 16K-token context, which needs a little more memory.

KV Cache + Overhead ≈ 3 GB (at full 262K context)

VRAM usage by quantization

Q4_K_M

21.2 GB

Q4_K_M + full context

23.8 GB

Learn more about VRAM estimation →

Can NVIDIA GeForce RTX 4090 run Qwen AgentWorld 35B A3B?

Yes, at Q4_K_M (21.2 GB) or lower. Higher quantizations like Q5_K_S (24.2 GB) exceed the NVIDIA GeForce RTX 4090's 24 GB.

What's the best quantization for Qwen AgentWorld 35B A3B?

For Qwen AgentWorld 35B A3B, Q4_K_M (21.2 GB) offers the best balance of quality and VRAM usage. Q5_K_S (24.2 GB) provides better quality if you have the VRAM. The smallest option is IQ2_XXS at 9.9 GB.

VRAM requirement by quantization

IQ2_XXS

9.9 GB~53%

Q2_K

15.1 GB~75%

IQ4_XS

19.0 GB~87%

Q4_K_M ★

21.2 GB~89%

Q5_K_S

24.2 GB~92%

BF16

69.7 GB~100%

★ Recommended — best balance of quality and VRAM usage.

Learn more about quantization →

Can I run Qwen AgentWorld 35B A3B on a Mac?

Yes — Mac Mini M2 Pro (16 GB) and 133 other Macs can run Qwen AgentWorld 35B A3B, the smallest at IQ2_M. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 21.2 GB of usable unified memory (RAM minus macOS overhead), from Mac Studio M1 Max (32 GB).

Can I run Qwen AgentWorld 35B A3B locally?

Yes — Qwen AgentWorld 35B A3B can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.

How fast is Qwen AgentWorld 35B A3B?

At Q4_K_M, Qwen AgentWorld 35B A3B can reach ~119 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~184 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.

tok/s = 1000 ÷ (active GB ÷ (bandwidth GB/s × efficiency) × 1000 + layers × routing ms)

Mixture-of-Experts: just the active experts get read for each token, along with a fixed routing cost per layer.

Example: NVIDIA B200 → 2.1 GB active ÷ (8000 × 0.65) = 0.406 ms, plus 40 layers × 0.055 ms = 2.200 ms, so 1000 ÷ 2.606 ms = ~384 tok/s

Estimated speed at Q4_K_M (21.2 GB)

NVIDIA B200

~384 tok/s

NVIDIA GeForce RTX 4090

~184 tok/s

NVIDIA B300

~384 tok/s

NVIDIA GH200 Grace Hopper Superchip

~349 tok/s

Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.

Learn more about tok/s estimation →

What's the download size of Qwen AgentWorld 35B A3B?

At Q4_K_M, the download is about 20.80 GB. The full-precision BF16 version is 69.32 GB. The smallest option (IQ2_XXS) is 9.53 GB.

Which GPUs can run Qwen AgentWorld 35B A3B?

7 consumer GPUs can run Qwen AgentWorld 35B A3B at Q4_K_M. Top options include NVIDIA GeForce RTX 5090, AMD Radeon RX 7900 XTX, NVIDIA GeForce RTX 3090. 1 GPU has plenty of headroom for comfortable inference.

Which devices can run Qwen AgentWorld 35B A3B?

100 devices with unified memory can run Qwen AgentWorld 35B A3B at Q4_K_M (21.2 GB), including AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10, Beelink GTR9 Pro (Ryzen AI Max+ 395, 128 GB), Framework Desktop (Ryzen AI Max+ 395, 128 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.