On November 5 Ant Group’s Ant Ling (@AntLingAGI) launched Ling-3.2-flash: a 560B-total / 25B-active native multimodal MoE with a 1M context window and native image, text and video input. The official changelog targets complex visual understanding, multimodal long-horizon agents, 3D understanding and interaction, and medical multimodal work; APIs are OpenAI- and Anthropic-compatible and free for the first two weeks. The launch post cites vendor evals versus Claude Opus 5 (max): OmniDocBench1.5 92.21 vs 90.56, ZEROBench(Sub) 43.71 vs 39.52, Vision2Web 85.86 vs 80.20. DesignArena placed it about #10 on Image-to-Frontend with Elo 1238.
Key Takeaways
- ✓Specs (official docs): 560B total / 25B active; 1M context; native image/text/video; four focuses: complex vision, multimodal long-horizon agents, 3D, medical multimodal
- ✓Availability: chat UI plus OpenAI/Anthropic-compatible APIs; changelog says free for first 2 weeks, then original pricing resumes
- ✓Vendor evals (official X): OmniDocBench1.5 92.21 vs Opus 5 90.56; ZEROBench(Sub) 43.71 vs 39.52; Vision2Web 85.86 vs 80.20
- ✓Third party: DesignArena Image-to-Frontend ~#10, Elo 1238; official price page stresses 2-week free promo and has not listed 3.2 USD rates yet
- ✓Hands-on: model id
Ling-3.2-flash; use image_url / video_url blocks; weights not on Hugging Face (API-first)
Key Decision Metrics at a Glance
Turn your technical choice into a development budget
Compare 40 dev plans & simulate token costs vs $20/mo subscriptions
Project Links & Resources
Direct AccessIn-Depth Technical Analysis
Core Background & Industry Pain Points
Multimodal agents need screenshots, PDFs and video as first-class inputs. After Ling-3.1, Ant Ling launched Ling-3.2-flash on November 5, folding vision into a 560B/25B MoE for image-to-site, document understanding and multimodal long-horizon work.
Architecture Highlights & Internals
Official docs: native multimodal MoE, 560B total / ~25B active, 1M context, native image/text/video. Focuses: complex vision, multimodal long-horizon agents, 3D and medical multimodal. API id Ling-3.2-flash with image_url/video_url, OpenAI and Anthropic compatible.
Authoritative Benchmarks & Measured Scores
Vendor X post versus Claude Opus 5 (max): OmniDocBench1.5 92.21 vs 90.56; ZEROBench(Sub) 43.71 vs 39.52; Vision2Web 85.86 vs 80.20. DesignArena cited ~#10 Image-to-Frontend, Elo 1238. Official pricing page stresses free for the first two weeks and has not yet listed standalone USD rates for 3.2.
Developer Hands-on Guide
Create a token on developer.ant-ling.com and call Ling-3.2-flash via OpenAI/Anthropic SDKs; try the chat UI first. Fit for lower-cost multimodal and Chinese workflows; API-first today (weights not on Hugging Face). Watch free-period quotas and when list prices return.
Benchmark side-by-side against alternatives, or calculate monthly token cost vs subscription break-even.
Discussion & Comments
0Sign in to join the discussion
Connect with AI developers to exchange benchmark insights.