XingChen-AGIXing4_0ForCausalLM

Xing4.0 29B A4B — Hardware Requirements & GPU Compatibility

ChatFunctions

Xing4.0 29B A4B is a 31.2B-parameter open language model from XingChen-AGI. It supports a context window of up to 262,144 tokens. At Q4_K_M it needs about 20.20 GB of VRAM — see which GPUs and Macs can run it below.

31.2B params · MoE5.5B active262K contextApache 2.0Sep 20265.2K downloads1.9K likes47.5K quant downloads

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 on 24 GB (Q4_K_M · ~132 tok/s · 8K)Macs from 24 GB (Q4_K_M · ~41 tok/s)

Apache 2.0

Reference GPU · RTX 4090 24 GB

Estimates, ±20% · how we estimate

Run Xing4.0 29B A4B

One recommended way. Every other method, OS and quant is in the step-by-step guide.Each way opens a step-by-step guide for your OS and quant.

Ways to runOther ways

each opens step-by-step

Run in cloud

Your own private copy · Q4_K_M · 32 GB GPU

The easiest way to try this model
Running in a few minutes, easy setup
Chat right here or connect any app

$5+ bonus for new users

Run Xing4.0 29B A4B

28.4 GB VRAM at 16K context

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/Venastine-Research/Xing4.0-29B-A4B-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/Venastine-Research/Xing4.0-29B-A4B-GGUF, then pick Q4_K_M. Or use the CLI (run LM Studio once first):
$lms get https://huggingface.co/Venastine-Research/Xing4.0-29B-A4B-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 Venastine-Research/Xing4.0-29B-A4B-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/Venastine-Research/Xing4.0-29B-A4B-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 Venastine-Research/Xing4.0-29B-A4B-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 XingChen-AGI/Xing4.0-29B-A4B --max-model-len 16384
Runs the original weights, not the Q4_K_M GGUF; needs ~69 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 TokenAI-zer/Xing4.0-29B-A4B-4bit-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 TokenAI-zer/Xing4.0-29B-A4B-4bit-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.
$Venastine-Research/Xing4.0-29B-A4B-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
A40 (48 GB)CheapestCheapest19 GB free	~100 tok/s	$0.97	Deploy on RunPodfrom $0.35/hr (opens in new tab)	48 GB · 19 GB free · ~100 tok/s · $0.97 / 1M tokens
RTX A6000 (48 GB)19 GB free	~108 tok/s	$1.11	Deploy on Vast.aifrom $0.43/hr (opens in new tab)	48 GB · 19 GB free · ~108 tok/s · $1.11 / 1M tokens
A100 40GB11 GB free	~176 tok/s	$0.96	Deploy on Vast.aifrom $0.61/hr (opens in new tab)	11 GB free · ~176 tok/s · $0.96 / 1M tokens
RTX 6000 Ada (48 GB)19 GB free	~127 tok/s	$1.41	Deploy on Vast.aifrom $0.65/hr (opens in new tab)	48 GB · 19 GB free · ~127 tok/s · $1.41 / 1M tokens
A40 (48 GB)19 GB free	~100 tok/s	$1.83	Deploy on Vast.aifrom $0.66/hr (opens in new tab)	48 GB · 19 GB free · ~100 tok/s · $1.83 / 1M tokens
RTX 6000 Ada (48 GB)19 GB free	~127 tok/s	$1.61	Deploy on RunPodfrom $0.74/hr (opens in new tab)	48 GB · 19 GB free · ~127 tok/s · $1.61 / 1M tokens
L40S (48 GB)19 GB free	~118 tok/s	$1.86	Deploy on RunPodfrom $0.79/hr (opens in new tab)	48 GB · 19 GB free · ~118 tok/s · $1.86 / 1M tokens
A100 80GB51 GB free	~206 tok/s	$1.88	Deploy on RunPodfrom $1.39/hr (opens in new tab)	51 GB free · ~206 tok/s · $1.88 / 1M tokens

Compare GPUs →

Your own private copy of Xing4.0 29B A4B · Q4_K_M · OpenAI-compatible API

Open the llama.cpp listing on RunPod
Open on RunPodServerless (opens in new tab)
New to RunPod? Sign up and add credit first.
2
Fill in the form
Press Deploy and fill in:
Model
Venastine-Research/Xing4.0-29B-A4B-GGUF
Quantization
Q4_K_M
Context Size
16384
After pasting the model press Esc — don’t pick a suggestion. Press Next.
Create the endpoint
Untick 24 GB, keep 48 GB, then Create Endpoint. Right away: Manage → Edit endpoint → Max workers 1 → Save (changing settings later restarts the worker).
The first start downloads the model, usually 5–15 min. After that, replies after a pause come in seconds.
Create an API key
RunPod → Settings → API Keys → Create. Pick “Restricted” with access to this endpoint only.

Use it

Copy the endpoint URL from its page on RunPod (any of its links works) and paste it with the key. Both stay in this browser.

Endpoint URL

API key

Chats and keys stay only in this browser.

https://api.runpod.ai/v2/<ENDPOINT_ID>/openai/v1

Put your endpoint ID in place of <ENDPOINT_ID>. The API key is the one from step 2.

from openai import OpenAI
client = OpenAI(base_url="https://api.runpod.ai/v2/<ENDPOINT_ID>/openai/v1", api_key="rpa_...")
r = client.chat.completions.create(
    model="any",
    messages=[{"role": "user", "content": "Hello!"}])
print(r.choices[0].message.content)

Works with any OpenAI client.

https://api.runpod.ai/v2/<ENDPOINT_ID>/openai/v1

Put your endpoint ID in place of <ENDPOINT_ID>. The API key is the one from step 2.

Free on Windows, macOS, Linux, iOS, Android and the web.

Open the web app ↗ (opens in a new tab)Download ↗ (opens in a new tab)

1. Settings → Model Provider → Add → Add Custom Provider
2. API Mode OpenAI API Compatible; paste the API key and the base URL as API Host
3. Model → Fetch, pick the model
4. New chat → pick the provider → start chatting

Free and open source for Windows, macOS and Linux.

Download Jan ↗ (opens in a new tab)

1.Settings → next to Model Providers press +
2.API format OpenAI-compatible; Base URL is the base URL above (with /v1); API key is your key; Create
3.If no model appears, click + in Models and type any
4.New chat → pick the model → start chatting

How Much VRAM Does Xing4.0 29B A4B Need?

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

VRAM	+32K	+128K	File	Quality	On RTX 4090 (reference)
12.0 GB	29.6 GB	86 GB	9.9 GB	2-bit · i-quant medium — Importance-weighted 2-bit, medium	Fits: 16K context · ~185 tok/s
13.6 GB	31.2 GB	87.6 GB	11.6 GB	3-bit · i-quant — Importance-weighted 3-bit	Fits: 16K context · ~172 tok/s
14.7 GB	32.4 GB	88.7 GB	~13.3 GB	2-bit · K-quant — 2-bit quantization with K-quant improvements	Fits: 16K context · ~163 tok/s
15.5 GB	33.1 GB	89.5 GB	13.1 GB	3-bit · i-quant medium — Importance-weighted 3-bit, medium	Fits: 16K context · ~158 tok/s
16.7 GB	34.3 GB	90.7 GB	~15.2 GB	3-bit medium — 3-bit medium quantization	Fits: 8K context · ~150 tok/s
20.2 GB	37.8 GB	94.2 GB	19 GB	4-bit medium — 4-bit medium quantization — most popular sweet spot	Fits: 8K context · ~132 tok/s
23.7 GB	41.3 GB	97.7 GB	22.2 GB	5-bit medium — 5-bit medium quantization — good quality/size tradeoff	Fits: 2K context · ~117 tok/s
27.2 GB	44.9 GB	101.2 GB	25.7 GB	6-bit, very good — 6-bit quantization, very good quality	Does not fit: needs 35.5 GB
32.7 GB	50.3 GB	106.7 GB	33.2 GB	8-bit, near-lossless — 8-bit quantization, near-lossless	Does not fit: needs 40.9 GB
63.9 GB	81.5 GB	137.9 GB	62.5 GB	16-bit · bfloat — Brain floating point 16 — preferred for training	Does not fit: needs 72.1 GB

est.calculated, no published file yet

Selected Q4_K_M

4-bit medium quantization

28.4GBat 16K context

Weights: 18.7 GB
KV cache: 9.4 GB
Overhead: 0.3 GB

Fits RTX 4090 (reference) · fits up to 8K context · ~132 tok/s

≥50 tok/s needs ≥305 GB/s · RTX 5090 ~191, M3 Ultra ~71

Which Hardware Can Run Xing4.0 29B A4B?

Q4_K_M · 20.2 GB

Xing4.0 29B A4B (Q4_K_M) requires 20.2 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 27+ GB is recommended. Using the full 262K context window can add up to 149.2 GB, bringing total usage to 169.3 GB. 7 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti. 122 devices with unified memory can run Xing4.0 29B A4B, 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

8–16 GB · RTX 4060, RTX 3060 12GB, RTX 5060 Ti 16GBDoesn't fit — smallest quant needs 20.2 GB8 GB tier →12 GB tier →16 GB tier →

24 GB · RTX 4090referenceQ4_K_M · 28.4 GB · ~132 tok/s · 8Ktight28.4~1328K24 GB tier →

Mac 32 GB · Mac Mini M4Q4_K_M · 28.4 GB · ~20 tok/s · 16K28.4~2016KMacs →

Mac 64 GB · Mac Studio M4 MaxQ4_K_M · 28.4 GB · ~58 tok/s · 64K28.4~5864KMacs →

Alternatives to Xing4.0 29B A4B

Similar models

6tok/s on RTX 4090

Model	Score	VRAM	tok/s	Why it is similar
Hy Memory 30B A3BTencent	—	19.1 GB	~184	same size
Granite Switch 4.1 30BIBM	—	22.7 GB	~35	same size
Muse Glimmer 30BMeta	—	19.6 GB	~36	same size
NVIDIA Nemotron 3.5 Lightning 30B A3B BF16NVIDIA	—	19.3 GB	~197	same size
Hy MT2 30B A3BTencent	—	19.1 GB	~170	same size
Gemma 4 31BGoogle	—	21.7 GB	~32	same size

Fine-tunes & variants(5)community builds of Xing4.0 29B A4B

Specifications

Publisher: XingChen-AGI
Parameters: 31.2B
Active parameters: 5.5B per token
Architecture: Xing4_0ForCausalLM
Context Length: 262,144 tokens
Vocabulary Size: 131,072
Release Date: 2026-09-16
License: Apache 2.0

Resources

Hugging Face model card (opens in a new tab)

Frequently Asked Questions

How much VRAM does Xing4.0 29B A4B need?

Xing4.0 29B A4B requires 20.2 GB of VRAM at Q4_K_M, or 63.9 GB at BF16. Full 262K context adds up to 149.2 GB (169.3 GB total).

VRAM = Weights + KV Cache + Overhead

Weights = 31.2B × 4.8 bits ÷ 8 = 18.7 GB

KV Cache + Overhead ≈ 1.5 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 ≈ 150.7 GB (at full 262K context)

VRAM usage by quantization

Q4_K_M

20.2 GB

Q4_K_M + full context

169.3 GB

Learn more about VRAM estimation →

Can NVIDIA GeForce RTX 4090 run Xing4.0 29B A4B?

Yes, at Q5_K_M (23.7 GB) or lower. Higher quantizations like Q6_K (27.2 GB) exceed the NVIDIA GeForce RTX 4090's 24 GB.

What's the best quantization for Xing4.0 29B A4B?

For Xing4.0 29B A4B, Q4_K_M (20.2 GB) offers the best balance of quality and VRAM usage. Q5_K_M (23.7 GB) provides better quality if you have the VRAM. The smallest option is IQ2_M at 12.0 GB.

VRAM requirement by quantization

IQ2_M

12.0 GB~62%

Q2_K

14.7 GB~75%

Q4_K_M ★

20.2 GB~89%

Q5_K_M

23.7 GB~92%

Q6_K

27.2 GB~95%

BF16

63.9 GB~100%

★ Recommended — best balance of quality and VRAM usage.

Learn more about quantization →

Can I run Xing4.0 29B A4B on a Mac?

Yes — Mac Mini M5 Pro (24 GB) and 106 other Macs can run Xing4.0 29B A4B. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 20.2 GB of usable unified memory (RAM minus macOS overhead).

Can I run Xing4.0 29B A4B locally?

Yes — Xing4.0 29B A4B can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.

How fast is Xing4.0 29B A4B?

At Q4_K_M, Xing4.0 29B A4B can reach ~115 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~132 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 → 3.5 GB active ÷ (8000 × 0.65) = 0.68 ms, plus 40 layers × 0.055 ms = 2.20 ms, so 1000 ÷ 2.88 ms = ~347 tok/s

Estimated speed at Q4_K_M (20.2 GB)

NVIDIA B200

~347 tok/s

NVIDIA GeForce RTX 4090

~132 tok/s

NVIDIA B300

~347 tok/s

NVIDIA GH200 Grace Hopper Superchip

~302 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 Xing4.0 29B A4B?

At Q4_K_M, the download is about 18.73 GB. The full-precision BF16 version is 62.43 GB. The smallest option (IQ2_M) is 10.54 GB.

Which GPUs can run Xing4.0 29B A4B?

7 consumer GPUs can run Xing4.0 29B A4B 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 Xing4.0 29B A4B?

122 devices with unified memory can run Xing4.0 29B A4B at Q4_K_M (20.2 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.