WeVisOne VL 2B — Hardware Requirements & GPU Compatibility
VisionFunctions
WeVisOne-VL-2B is the smaller of two sizes of WeVisOne-VL, a unified vision-language model from WeChat Vision at Tencent, built on Qwen3-VL. One model covers object detection, referring localization, counting, pointing and OCR, as well as referring segmentation and promptable alpha matting, using structured coordinate tokens and a 9.6M-parameter pixel decoder for the pixel-level outputs. Context length is 262,144 tokens. The code is released under the Apache 2.0 license, and third-party components keep their own licenses. It is meant for single-turn image perception research, not video or chat. At about 2 billion parameters it is the lighter choice next to the 4B version and should run on a laptop or entry-level GPU, using the project's inference package for the mask outputs.
Runs great on a 24 GB GPU
Runs from 8 GB (Q4_K_M · 32K)Best on 24 GB (Q4_K_M · ~328 tok/s · 128K)Macs from 8 GB (Q4_K_M · ~36 tok/s)
Apache 2.0
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run WeVisOne VL 2B
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.
Recommended
ollama run hf.co/mradermacher/WeVisOne-VL-2B-GGUF:Q4_K_MQ4_K_M fits fully on the RTX 4090 · 20.4 GB free · 16K context
Download: WeVisOne-VL-2B.Q4_K_M.gguf1.28 GBmradermacher
Ways to runOther ways
each opens step-by-stepRun in cloud
Runs on RTX 3060 12GB (8 GB headroom) · Q4_K_M
- Generation speed
- ~117 tok/s
- generation speed
- Cost per 1M output tokens
- $0.15
- per 1M output tokens
How Much VRAM Does WeVisOne VL 2B Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 1.6 GB | 5.1 GB | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Fits: 128K context · ~417 tok/s | |
| 1.6 GB | 5.1 GB | 3-bit small — 3-bit small quantization | Fits: 128K context · ~410 tok/s | |
| 1.7 GB | 5.3 GB | 3-bit medium — 3-bit medium quantization | Fits: 128K context · ~379 tok/s | |
| 1.8 GB | 5.3 GB | 3-bit large — 3-bit large quantization | Fits: 128K context · ~366 tok/s | |
| 2 GB | 5.5 GB | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Fits: 128K context · ~328 tok/s | |
| 2.3 GB | 5.8 GB | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Fits: 128K context · ~289 tok/s | |
| 2.5 GB | 6.1 GB | 6-bit, very good — 6-bit quantization, very good quality | Fits: 128K context · ~257 tok/s | |
| 3.0 GB | 6.5 GB | 8-bit, near-lossless — 8-bit quantization, near-lossless | Fits: 128K context · ~220 tok/s |
Selected Q4_K_M
4-bit medium quantization3.6GBat 16K context
- Weights
- 1.5 GB
- KV cache
- 1.9 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · 20.4 GB free · ~328 tok/s
≥50 tok/s needs ≥154 GB/s · RTX 5090 ~582, RTX 3090 Ti ~328
Which Hardware Can Run WeVisOne VL 2B?
Q4_K_M · 2 GBWeVisOne VL 2B (Q4_K_M) requires 2 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 3+ GB is recommended. Using the full 262K context window can add up to 29.8 GB, bringing total usage to 31.8 GB. 52 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti. 163 devices with unified memory can run WeVisOne VL 2B, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB).
≤ 11 GB
≤ 11 GB tier →fits · 4.4 GB free · up to 32K ctx
16 GPUs · 1 Mac class · 2 device families
RTX 3080~247Simulate output on RTX 3080RTX 2080 Ti~200Simulate output on RTX 2080 TiRTX 3070 Ti~198Simulate output on RTX 3070 TiM3*8 GB~36Simulate output on iMac 24" M3 (8 GB)Orin Nano~33Simulate output on NVIDIA Jetson Orin Nano 8GB (Super)iPhone~24Simulate output on iPhone 1720 GB
fits · 16.4 GB free · up to 128K ctx
1 GPU
RX 7900 XT~240Simulate output on RX 7900 XT32–40 GB
fits · 28.4 GB free · up to 256K ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~582Simulate output on RTX 5090M5 Max36 GB~161Simulate output on Mac Studio M5 Max (36 GB)M532 GB~54Simulate output on MacBook Air 13" M5 (32 GB)Orin~67Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~44Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at Q4_K_M
Similar models
6| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| MiniCPM5 2BOpenBMB | — | 2.5 GB | ~345 | same size |
| North Micro Vision InstructCohere | — | 3.7 GB | ~323 | same size |
| Qwen3.5 2BAlibaba | — | 1.9 GB | ~370 | same size |
| Granite 3.1 2B InstructIBM | — | 3.2 GB | ~329 | same size |
| Gemma 2 2B ITGoogle | — | 2.5 GB | ~379 | same size |
| EXAONE 3.5 2.4B InstructLGAI-EXAONE | — | 3.4 GB | ~412 | same size |
Specifications
- Publisher
- Tencent
- Parameters
- 2.4B
- Architecture
- Qwen3VLForConditionalGeneration
- Context Length
- 262,144 tokens
- Vocabulary Size
- 152,680
- Release Date
- 2026-10-22
- License
- Apache 2.0
Frequently Asked Questions
How much VRAM does WeVisOne VL 2B need?
WeVisOne VL 2B requires 2 GB of VRAM at Q4_K_M, or 5.4 GB at BF16. Full 262K context adds up to 29.8 GB (31.8 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 2.4B × 4.8 bits ÷ 8 = 1.5 GB
KV Cache + Overhead ≈ 0.5 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 ≈ 30.3 GB (at full 262K context)
VRAM usage by quantization
What's the best quantization for WeVisOne VL 2B?
For WeVisOne VL 2B, Q4_K_M (2 GB) offers the best balance of quality and VRAM usage. Q5_K_S (2.2 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 1.6 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run WeVisOne VL 2B on a Mac?
Yes — MacBook Air 13" M3 (8 GB) and 145 other Macs can run WeVisOne VL 2B. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 2 GB of usable unified memory (RAM minus macOS overhead).
Can I run WeVisOne VL 2B locally?
Yes — WeVisOne VL 2B can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is WeVisOne VL 2B?
At Q4_K_M, WeVisOne VL 2B can reach ~2400 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~328 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = (bandwidth GB/s ÷ model GB) × efficiency
Example: NVIDIA B200 → 8000 ÷ 2.0 × 0.65 = ~2600 tok/s
Estimated speed at Q4_K_M (2 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of WeVisOne VL 2B?
At Q4_K_M, the download is about 1.47 GB. The full-precision BF16 version is 4.88 GB. The smallest option (Q2_K) is 1.04 GB.
Which GPUs can run WeVisOne VL 2B?
52 consumer GPUs can run WeVisOne VL 2B at Q4_K_M. Top options include AMD Radeon RX 6700 XT, AMD Radeon RX 6800, AMD Radeon RX 6800 XT. 52 GPUs have plenty of headroom for comfortable inference.
Which devices can run WeVisOne VL 2B?
163 devices with unified memory can run WeVisOne VL 2B at Q4_K_M (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.