You can deploy this model and start earning money today!

mlx-community/
UI-Venus-1.5-8B-bf16

Name: UI-Venus-1.5-8B-bf16
Availability: InStock

Views

No views yet

UI-Venus-1.5-8B bf16

This is an MLX conversion of inclusionAI/UI-Venus-1.5-8B, optimized for Apple Silicon.

UI-Venus-1.5 is a unified end-to-end GUI agent family built for grounding, web navigation, and mobile navigation. The 1.5 family spans dense 2B and 8B variants plus a 30B-A3B MoE variant, and is framed upstream around a shared GUI semantics stage, online RL for long-horizon navigation, and model merging across grounding, web, and mobile domains.

This MLX artifact was converted with mlx-vlm and validated locally with both mlx_vlm prompt-packet checks and vllm-mlx OpenAI-compatible serve checks.

Conversion Details

Field	Value
Upstream model	`inclusionAI/UI-Venus-1.5-8B`
Artifact type	`bf16 MLX conversion`
Conversion tool	`mlx_vlm.convert` via `mlx-vlm 0.3.12`
Python	`3.11.14`
MLX	`0.31.0`
Transformers	`5.2.0`
Validation backend	`vllm-mlx (phase/p1 @ 8a5d41b)`
Quantization	`bf16`
Group size	`n/a`
Quantization mode	`n/a`
Artifact size	`16.34G`
Template repair	`tokenizer_config.json["chat_template"]` was re-injected after conversion

Additional notes:

Upstream UI-Venus ships both chat_template.json and tokenizer_config.json["chat_template"].
This MLX artifact preserves the dual-template contract across chat_template.json, chat_template.jinja, and tokenizer_config.json["chat_template"].
chat_template.jinja is present as an additive compatibility shim for downstream tooling.

Validation

This artifact passed local validation in this workspace:

mlx_vlm prompt-packet validation: PASS
vllm-mlx OpenAI-compatible serve validation: PASS

Local validation notes:

structured action was strong on the fixed packet and retained the requested reason field
the main caution was grounding looseness rather than text or schema degeneration

Performance

Artifact size on disk: 16.34G
Local fixed-packet mlx_vlm validation used about 30.55 GB peak memory
Observed local fixed-packet throughput was about 206-244 prompt tok/s and 20.4-26.8 generation tok/s across the four validation prompts
Local vllm-mlx serve validation passed cleanly on both non-stream and streamed multimodal chat

These are local validation measurements, not a full benchmark suite.

Usage

Install

pip install -U mlx-vlm

CLI

bash

python -m mlx_vlm.generate \
  --model mlx-community/UI-Venus-1.5-8B-bf16 \
  --image path/to/image.png \
  --prompt "Describe the visible controls on this screen." \
  --max-tokens 256 \
  --temperature 0.0

Python

python

from mlx_vlm import load, generate

model, processor = load("mlx-community/UI-Venus-1.5-8B-bf16")
result = generate(
    model,
    processor,
    prompt="Describe the visible controls on this screen.",
    image="path/to/image.png",
    max_tokens=256,
    temp=0.0,
)
print(result.text)

vllm-mlx Serve

python -m vllm_mlx.cli serve mlx-community/UI-Venus-1.5-8B-bf16 --mllm --localhost --port 8000

Other Quantizations

Planned sibling repos in this wave:

Notes and Limitations

This card reports local MLX conversion and validation results only.
Upstream benchmark claims belong to the original UI-Venus model family and were not re-run here unless explicitly stated.
This artifact is the local reference artifact for the 6bit and 4bit variants in this wave.
On the fixed local packet, the main qualitative caution was loose grounding rather than schema breakage.

Citation

If you use this MLX conversion, please cite the original UI-Venus papers:

License

This repo follows the upstream model license: Apache 2.0. See the upstream model card for the authoritative license details: inclusionAI/UI-Venus-1.5-8B.

mlx-community/UI-Venus-1.5-8B-bf16