Jan-v2-VL: Multimodal Agent for Long-Horizon Tasks
Overview
Jan-v2-VL-max
extends the Jan-v2-VL family to a
30B-parameter
vision–language model focused on
long-horizon execution
. This release scales model capacity and applies
LoRA-based RLVR
to improve stability over many steps with
low error accumulation
. For evaluation, we continue to use
The Illusion of Diminishing Returns: Measuring Long-Horizon Execution in LLMs
, which emphasizes execution length rather than knowledge recall.
Intended Use
Tasks where the plan and/or knowledge can be provided up front, and success hinges on stable, many-step execution with minimal drift:
Agentic automation & UI control:
Stepwise operation in browsers/desktop apps with screenshot grounding and tool calls via
Jan Browser MCP
.
Model Performance
Evaluated under FP8 inference,
Jan-v2-VL-max
vs.
Qwen3-VL-30B-A3B-Thinking
shows
no regressions
and
small gains
on several tasks, with the largest improvements in
long-horizon execution
. Our FP8 build maintains accuracy while reducing memory footprint and latency.
Local Deployment
Jan Web
Hosted on
Jan Web
— use the model directly at
chat.jan.ai
Local Deployment
Using vLLM:
We recommend
vLLM
for serving and inference. All reported results were run with
vLLM 0.12.0
.
For
FP8
deployment, we used
llm-compressor
built from source. Please pin transformers==4.57.1 for compatibility.
# Exact versions used in our evals
pip install vllm==0.12.0
pip install transformers==4.57.1
pip install "git+https://github.com/vllm-project/llm-compressor.git@1abfd9eb34a2941e82f47cbd595f1aab90280c80"
Jan-v2-VL-max-FP8 huggingface.co is an AI model on huggingface.co that provides Jan-v2-VL-max-FP8's model effect (), which can be used instantly with this janhq Jan-v2-VL-max-FP8 model. huggingface.co supports a free trial of the Jan-v2-VL-max-FP8 model, and also provides paid use of the Jan-v2-VL-max-FP8. Support call Jan-v2-VL-max-FP8 model through api, including Node.js, Python, http.
Jan-v2-VL-max-FP8 huggingface.co is an online trial and call api platform, which integrates Jan-v2-VL-max-FP8's modeling effects, including api services, and provides a free online trial of Jan-v2-VL-max-FP8, you can try Jan-v2-VL-max-FP8 online for free by clicking the link below.
janhq Jan-v2-VL-max-FP8 online free url in huggingface.co:
Jan-v2-VL-max-FP8 is an open source model from GitHub that offers a free installation service, and any user can find Jan-v2-VL-max-FP8 on GitHub to install. At the same time, huggingface.co provides the effect of Jan-v2-VL-max-FP8 install, users can directly use Jan-v2-VL-max-FP8 installed effect in huggingface.co for debugging and trial. It also supports api for free installation.