Jan-v2-VL
An 8B vision-language model that sees the screen and carries out long, many-step tasks in browsers and desktop apps with minimal drift.
Overview
- Parameters
- 8B
- Base model
- Qwen3-VL-8B-Thinking
- Modality
- Vision-language
- Variants
- Low, med, high
- License
- Apache 2.0
49
Long-horizon benchmark · task length (K)
Jan-v2-VL is an 8B vision-language model built on Qwen3-VL-8B-Thinking, designed for long-horizon, multi-step agentic tasks in real software environments — browsers, desktop apps, and beyond. It combines language reasoning with visual perception to execute stable, many-step automation chains with minimal drift.
Try it in JanVariants
Jan-v2-VL comes in three variants tuned for different latency/quality tradeoffs:
| Variant | Focus |
|---|---|
| Jan-v2-VL-low | Efficiency-oriented, lower latency |
| Jan-v2-VL-med | Balanced latency and quality |
| Jan-v2-VL-high | Deeper reasoning, higher think time |
Capabilities
- Agentic automation & UI control: Follow complex multi-step instructions in browsers and desktop apps
- Screenshot grounding: Perceive and act on visual state of the screen
- Tool calling: Native support for BrowserMCP and similar integrations
- Error recovery: Maintain intermediate state and recover from minor execution errors
- Long-horizon execution: Stable performance across many-step automation chains
Performance
Long-Horizon Execution
Jan-v2-VL dramatically outperforms competing VLMs on the long-horizon benchmark, completing tasks up to 49 steps in a single turn — nearly 10x more than the next best model.
| Benchmark | Jan-v2-VL | Qwen3-VL-8B-Thinking | Qwen2.5-VL-7B-Instruct | Gemma3-it-12B | Llama3.1-nemotron | GLM-4.1V-9B-Thinking |
|---|---|---|---|---|---|---|
| Long-horizon benchmark (Task length (K) executed in a single turn) | 49 | 5 | 2 | 2 | 1 | 1 |
All three Jan-v2-VL variants significantly exceed the Qwen3-VL-8B-Thinking baseline (score: 5):
| Benchmark | Jan-v2-VL-high | Jan-v2-VL-med | Jan-v2-VL-low | Qwen3-VL-8B-Thinking (baseline) |
|---|---|---|---|---|
| Long-horizon execution (Task length (K) executed in a single turn) | 49 | 37 | 29 | 5 |
Text-Only Benchmarks
Jan-v2-VL maintains strong performance on standard language benchmarks with no degradation from the base model:
| Benchmark | Qwen3-VL-8B-Thinking (baseline) | Jan-v2-VL-low | Jan-v2-VL-med | Jan-v2-VL-high |
|---|---|---|---|---|
| MMLU | 74.38 | 74.36 | 74.38 | 74.43 |
| SimpleQA | 87 | 88.3 | 87.1 | 87.1 |
| IF Bench – Prompt Acc. | 0.3401 | 0.3265 | 0.3231 | 0.3639 |
| IF Bench – Instr. Acc. | 0.3701 | 0.3612 | 0.3552 | 0.3851 |
| AIME25 | 0.80 | 0.76 | 0.67 | 0.87 |
| EQ Bench | 84.4 | 84.1 | 84.6 | 84.1 |
| Creative Writing | 79.1 | 75.07 | 73.0 | 74.0 |
| SWE Bench (✓ / 300) | 10 | 11 | 9 | 9 |
Multimodal Benchmarks
Core vision capabilities are fully preserved across all variants:
| Benchmark | Qwen3-VL-8B-Thinking (baseline) | Jan-v2-VL-low | Jan-v2-VL-med | Jan-v2-VL-high |
|---|---|---|---|---|
| RealVQA | 52.81 | 51.5 | 51.7 | 51.2 |
| CountBench | 88.79 | 88.79 | 88.18 | 88.79 |
| OCR bench v2 (EN) | 0.582 | 0.583 | 0.583 | 0.585 |
| OCR bench v2 (CN) | 0.595 | 0.595 | 0.588 | 0.609 |
Requirements
- Memory:
- Minimum: 16GB RAM (with Q4 quantization)
- Recommended: 24GB RAM (with Q8 quantization)
- Hardware: GPU recommended for real-time agentic use
- API Support: OpenAI-compatible at localhost:1337
Using Jan-v2-VL
Quick Start
- Download Jan Desktop
- Click Open in Jan above, or open the Hub and search for Jan-v2-VL
- Choose Jan-v2-VL-med or another variant that fits your hardware, then click Download
- Once downloaded, select the downloaded Jan-v2-VL variant from the model selector in any chat
- Start using it for browser automation or visual tasks
Deployment Options
Using vLLM:
vllm serve janhq/Jan-v2-VL-high \ --host 0.0.0.0 \ --port 1234 \ --enable-auto-tool-choice \ --tool-call-parser hermes \ --reasoning-parser qwen3
Using llama.cpp:
llama-server --model Jan-v2-VL-high-Q8_0.gguf \ --vision-model-path mmproj-Jan-v2-VL-high.gguf \ --host 0.0.0.0 \ --port 1234 \ --jinja \ --no-context-shift
Recommended Parameters
temperature: 1.0top_p: 0.95top_k: 20repetition_penalty: 1.0presence_penalty: 1.5
What Jan-v2-VL Does Well
- UI automation: Navigate and interact with browsers and desktop applications via screenshots
- Long-horizon tasks: Execute multi-step plans without losing context or drifting
- Visual grounding: Identify and interact with specific UI elements from screenshots
- Tool calling: Integrate with MCP servers like BrowserMCP for end-to-end automation
Limitations
- GPU recommended: Real-time agentic use benefits significantly from GPU acceleration
- 8B size: More capable than 4B models for visual tasks, but larger resource footprint
- Vision required: Text-only tasks do not benefit from the vision-language architecture
Downloads
- Jan: Download Jan
- Jan-v2-VL-low: Hugging Face (opens in a new tab) GGUF (opens in a new tab)
- Jan-v2-VL-med: Hugging Face (opens in a new tab) GGUF (opens in a new tab)
- Jan-v2-VL-high: Hugging Face (opens in a new tab) GGUF (opens in a new tab)
GGUF quantizations
- Q4_K_M: Good balance of size and quality
- Q5_K_M: Better quality, slightly larger
- Q8_0: Highest quality quantization
Talk to us
- Discussions: Hugging Face community (opens in a new tab)
- Chat: Discord (opens in a new tab)
- Issues: GitHub (opens in a new tab)



