Mage-VL-Cowork A Full Training Recipe to Learn Hybrid Computer Use from Evolved Harnesses
Alternatives to Mage-VL-Cowork
120 open-weight models that do the same job as Mage-VL-Cowork (4.7B vision), ranked by how close they are in task and size, then by popularity and how recent they are. Licence and VRAM are shown for each.
ZwZ2-4B is the Qwen3.5-4B-based model in Zooming-without-Zooming-v2 (ZwZ2). It learns to connect visual evidence across regions through comparison, aggregation, matching, and multi-step reasoning. Training uses full images; inference requires no zooming tools.
4.5B256K ctx Unknown licence1.1K 12Oct 2026- Same category
- Similar size (4.5B vs 4.7B)
LFM2.5-VL-3B is a multimodal variant of LFM2.5, a family of hybrid models designed for on-device deployment. It builds on LFM2-VL-3B with further mid- and post-training.
3.1B32K ctx lfm1.054.2K 239Aug 2026- Same category
- Similar size (3.1B vs 4.7B)
Precise grounding matters. It specifies which object is the target and where that object is, even in clutter and for tiny objects, and it has to be fast enough for closed-loop control.
4.8B256K ctx apache-2.0-with-upstream-kimi-k3956 44Sept 2026- Same category
- Similar size (4.8B vs 4.7B)
- Commercial use OK
We introduce InternVL3.5, a new family of open-source multimodal models that significantly advances versatility, reasoning capability, and inference efficiency along the InternVL series.
4.7B40K ctx Apache 2.016.2K 30Aug 2025- Same category
- Similar size (4.7B vs 4.7B)
- Commercial use OK
Gemma is a family of lightweight, state-of-the-art open models from Google, built from the same research and technology used to create the Gemini models.
4.3BGemma Terms of Use116K 180Feb 2025- Same category
- Similar size (4.3B vs 4.7B)
Important Update! We made a few small but important updates to this HF/transformers-compatible checkpoint to ensure exact outputs to our native model checkpoint on March 29, 2026 ~6PM PST. If you downloaded this model checkpoint earlier than this time, we recommend re-downloading it.
4.9B10K ctx Apache 2.02.4K 36Mar 2026- Same category
- Similar size (4.9B vs 4.7B)
- Commercial use OK
DeskForge-Qwen3.5-4B is Qwen3.5-4B fine-tuned for desktop GUI grounding and next-action prediction using grounding and GUI-workflow supervision.
4.5B256K ctx Apache 2.0623 9Sept 2026- Same category
- Similar size (4.5B vs 4.7B)
- Commercial use OK
d1-3B is a 3B parameter decision model built on LFM2.5-VL-3B. You give it a state (text, JSON, images, or a mix) and a set of questions. It returns calibrated, typed answers in one forward pass with zero output tokens.
3.1B32K ctx lfm1.019.1K 512Oct 2026- Same category
- Similar size (3.1B vs 4.7B)
In the past five months since Qwen2-VL’s release, numerous developers have built new models on the Qwen2-VL vision-language models, providing us with valuable feedback. During this period, we focused on building more useful vision-language models.
8.3B125K ctx Apache 2.05.2M 1.8KJan 2025- Same category
- Similar size (8.3B vs 4.7B)
- Commercial use OK
- Widely used
ZwZ-4B is a fine-grained multimodal perception model built upon Qwen3-VL-4B. It is trained using Region-to-Image Distillation (R2I) combined with reinforcement learning, enabling superior fine-grained visual understanding in a single forward pass — no inference-time zooming or tool calling…
4.8B256K ctx Apache 2.01.6K 34Feb 2026- Same category
- Similar size (4.8B vs 4.7B)
- Commercial use OK
A GPT-4V Level MLLM for Single Image, Multi Image and Video on Your Phone
4.1B32K ctx Apache 2.030.4K 465Jul 2025- Same category
- Similar size (4.1B vs 4.7B)
- Commercial use OK
Muse Glimmer is a 30-billion-parameter causal language model with a dedicated perception encoder, distilled from Muse Spark and purpose-built for autonomous agentic tasks on consumer hardware.
2.6B128K ctx Apache 2.090.3K 83Aug 2026- Same category
- Similar size (2.6B vs 4.7B)
- Commercial use OK
Phi-3.5-vision is a lightweight, state-of-the-art open multimodal model built upon datasets which include - synthetic data and filtered publicly available websites - with a focus on very high-quality, reasoning dense data both on text and vision.
4.1B128K ctx MIT237.6K 750Aug 2024- Same category
- Similar size (4.1B vs 4.7B)
- Commercial use OK
We introduce InternVL2.5-MPO, an advanced multimodal large language model (MLLM) series that demonstrates superior overall performance. This series builds upon InternVL2.5 and Mixed Preference Optimization.
3.7B32K ctx MIT245.9K 18Dec 2024- Same category
- Similar size (3.7B vs 4.7B)
- Commercial use OK
Model Summary: Granite-4.0-3B-Vision is a vision-language model (VLM) designed for enterprise-grade document data extraction. It focuses on specialized, complex extraction tasks that ultracompact models often struggle with:
4B128K ctx Apache 2.03.3K 113Mar 2026- Same category
- Similar size (4B vs 4.7B)
- Commercial use OK
ZAYA1-VL-8B is a vision-language model (VLM) built upon Zyphra's ZAYA1-8B LLM. It has state-of-the-art performance among VLMs for its size and inference efficiency.
9.7B32K ctx Apache 2.0339.4K 43Apr 2026- Same category
- Larger (9.7B vs 4.7B)
- Commercial use OK
This model is part of the GELab-Zero project, which presents the Step-GUI Technical Report [Paper] [Project Page] [Code].
4.4B256K ctx Apache 2.02.3K 153Nov 2025- Same category
- Similar size (4.4B vs 4.7B)
- Commercial use OK
NVIDIA Cosmos Reason 2 is an open, customizable, 2B-parameter reasoning vision language model (VLM) for physical AI and robotics that enables robots and vision AI agents to reason like humans, using prior knowledge, physics understanding and common sense to understand and act in the real world.
2.4B256K ctx NVIDIA Open Model License475.6K 420Dec 2025- Same category
- Similar size (2.4B vs 4.7B)
- Commercial use OK
Over recent months, we have intensified our focus on developing foundation models that deliver exceptional utility and performance.
2.3B256K ctx Apache 2.0321.5K 108Feb 2026- Same category
- Smaller (2.3B vs 4.7B)
- Commercial use OK
Over recent months, we have intensified our focus on developing foundation models that deliver exceptional utility and performance.
9.7B256K ctx Apache 2.0266.8K 163Feb 2026- Same category
- Larger (9.7B vs 4.7B)
- Commercial use OK
Youtu-VL is a lightweight yet robust Vision-Language Model (VLM) built on the Youtu-LLM with 4B parameters. It pioneers Vision-Language Unified Autoregressive Supervision (VLUAS), which markedly strengthens visual perception and multimodal understanding.
5.3B32K ctx youtu-vl1.4K 163Jan 2026- Same category
- Similar size (5.3B vs 4.7B)
MiMo-V2.6-Distill-Qwen-9B is a 9B agentic model developed by Xiaomi MiMo through supervised fine-tuning of Qwen3.5-9B on MiMo-generated data. It covers coding, general-purpose agent tasks, visual coding, and cybersecurity.
9.4B256K ctx MIT38.9K 715Sept 2026- Same category
- Similar size (9.4B vs 4.7B)
- Commercial use OK
FLUX 3 Image Klein Prompt Upsampler is a 9B vision-language model that expands a short image request into the detailed prompt FLUX 3 Image Klein works best with. It takes the request, up to 10 reference images and an optional list of bounding boxes, and returns one prompt.
9.4B256K ctx Apache 2.031K 83Oct 2026- Same category
- Similar size (9.4B vs 4.7B)
- Commercial use OK
Clef-Flash is a 9B multimodal model that turns a state and a schema of typed questions into decisions. It reads the state as text, JSON, images, or video, and returns a probability for every allowed option of every question in a single forward pass.
9.4B256K ctx Apache 2.032.7K 934Sept 2026- Same category
- Similar size (9.4B vs 4.7B)
- Commercial use OK