inclusionAI · Released, not on our provider

Ling 3.2 Flash (2026): inclusionAI’s 560B native multimodal model, specs and price

Ling 3.2 Flash is Ant Ling’s native multimodal model (listed on OpenRouter under inclusionAI), launched on November 5, 2026 with 560B total and 25B active parameters, a 1M-token context and image, text and video input; Ant Ling priced its first two weeks at zero, and OpenRouter lists a $0 variant.

Released, not on our providerLast updated:

Use it now

Chat

This chat starts with Inkling 2 Small, not Ling 3.2 Flash.

≈ 1 credit per reply · See all models

New accounts get 5 free credits once, valid for 7 days. They work on the lower-priced models here; the priciest ones need credits you bought.

Start chatting — 5 free credits
Maker
inclusionAI
Type
Text
Released
2026-11-05
OpenRouter id
None

Checked

Ling 3.2 Flash: facts, with sources

  • Ant Ling’s changelog lists Ling-3.2-flash under November 5, 2026 (platform version v2.0.6): launched as a native multimodal model with 560B total and 25B activated parameters, a 1M context window and native image, text and video input, available in the platform’s chat interface and through OpenAI- and Anthropic-compatible API calls.Source: developer.ant-ling.com · checked 2026-11-19
  • Ant Ling’s model page says Ling-3.2-flash fuses vision and language at the architecture level, so image, text and video input do not rely on external vision modules; it keeps 560B of capacity while activating about 25B per step, and its 1M context is meant for long documents and long video.Source: developer.ant-ling.com · checked 2026-11-19
  • Ant Ling names four focus areas for Ling-3.2-flash: complex visual understanding, multimodal long-horizon agents, 3D understanding and interaction, and medical multimodal work (medical imaging understanding and report interpretation to assist professional analysis).Source: developer.ant-ling.com · checked 2026-11-19
  • Ant Ling’s pricing page says Ling-3.2-flash is free for the first 2 weeks after launch and that original prices resume after the promotion; its price table lists no price for Ling-3.2-flash (it lists Ling-3.1-flash, Ling-3.0-flash-VL and Ling-3.0-flash). Before a first API call the page asks you to bind an Alipay account and activate the service on its console.Source: developer.ant-ling.com · checked 2026-11-19
  • The same pricing page adds that during the promotion the same models on OpenRouter and ZenMux are also discounted, and that original prices resume afterwards. It gives no end date; counted from the November 5 launch, two weeks end around November 19, 2026, the day we read the page (our arithmetic). The page says it was last updated on November 5, 2026.Source: developer.ant-ling.com · checked 2026-11-19
  • OpenRouter lists Ling 3.2 Flash only as inclusionai/ling-3.2-flash:free (no paid id on the list we read) at $0 per million input and output tokens: 1,048,576-token context, up to 262,144 completion tokens, text and image input, text output, one host (Ant Digital, fp8). OpenRouter’s page shows a release date of November 13, 2026.Source: openrouter.ai · checked 2026-11-19 · second-hand
  • OpenRouter’s page says the model is a hybrid reasoning model whose thinking can be switched off per request, that it takes up to 200 images per request, and that it supports tools with tool_choice and structured outputs through a JSON schema. The page also says free endpoints are rate limited.Source: openrouter.ai · checked 2026-11-19 · second-hand
  • OpenRouter’s limits page says free model variants (ids ending in :free) have per-minute and per-day request limits that depend on the credits an account has bought. Neither OpenRouter nor Ant Ling states how long the :free listing of Ling 3.2 Flash will stay.Source: openrouter.ai · checked 2026-11-19 · second-hand
  • Ant Ling’s model page describes the Ling series as independently developed and open-sourced by Ant Group; its Ling-3.2-flash section does not mention weights or a licence.Source: developer.ant-ling.com · checked 2026-11-19
  • Hugging Face’s list of inclusionAI models matching "Ling-3" (November 19, 2026) shows repositories for Ling-3.0 and Ling-3.1-flash but none for Ling-3.2, and the id OpenRouter gives (inclusionAI/Ling-3.2-flash) did not load when we requested it, so we found no public weights for Ling 3.2 Flash.Source: huggingface.co · checked 2026-11-19
  • AICoder reports that Ant Ling’s launch post on X (which we could not open) compares Ling-3.2-flash with Claude Opus 5 (max) in the maker’s own release evaluations: OmniDocBench1.5 92.21 vs 90.56, ZEROBench (Sub) 43.71 vs 39.52 and Vision2Web 85.86 vs 80.20. These are the maker’s figures as relayed by a third party; we have not tested them.Source: aicoder.com · checked 2026-11-19 · second-hand

Ling models

Ling 3.2 Flash next to Ling 3.1 Flash and the 3.0 models

  • Ant Ling’s model-selection list gives Ling-3.2-flash and Ling-3.1-flash a 1M context and 560B total / 25B activated parameters, Ling-3.0-flash-VL 256K and 124B / 5.5B, Ling-3.0-flash 256K and 124B / 5.1B, and Ling-3.0-tiny 7.9B / 1.3B with no context figure.Source: developer.ant-ling.com · checked 2026-11-19
  • Ant Ling positions Ling-3.1-flash as a hybrid reasoning model for everyday office work, long-horizon software development and complex agent tasks, and Ling-3.0-flash-VL as a native multimodal model with a visual feedback loop for image-to-web and GUI automation.Source: developer.ant-ling.com · checked 2026-11-19

Side by side

Ling 3.2 Flash: Ling-3.2-flash, 3.1-flash, 3.0-flash-VL and 3.0-flash compared

Ling 3.2 Flash is the newest model in Ant Group’s Ling series, and the table shows how Ant Ling itself separates it from the three models before it. It has the same size as Ling 3.1 Flash, 560B total and 25B active, with the same 1M context; the difference Ant Ling stresses is native image, text and video input and a focus on visual and medical work. Ant Ling’s own price table has no row for it, only the promotion, so the one number you can compare today is OpenRouter’s $0 for the :free variant, which is a promotional price and not a long-term one. This site’s chat does not run :free variants, so you cannot try Ling 3.2 Flash here. For a model with text and image input and a 1M-token context that you can try on this site, see Inkling 2 Small; for another multimodal Chinese model with a 1M context and a very low price, see Hy4 Flash; and Gemini 4 Hydrogen is a Google model with image input and the same 1,048,576-token window. Context and price figures for the older Ling models are the ones OpenRouter lists today, so they can differ from the numbers on Ant Ling’s own pages.

Our table, from Ant Ling’s model and pricing pages, OpenRouter’s model list and Hugging Face, all read on November 19, 2026 · checked 2026-11-19
WhatLing-3.2-flashLing-3.1-flashLing-3.0-flash-VLLing-3.0-flash
How Ant Ling describes itNative multimodal (image, text, video input): visual understanding, multimodal agents, 3D, medicalHybrid reasoning for office work, long-horizon development and agent tasksNative multimodal hybrid reasoning with a visual feedback loop (image-to-web, GUI automation)Cost-effective hybrid reasoning, high speed (peak up to 1000 tokens/s)
Context (Ant Ling)1M1M256K256K
Total / activated parameters560B / 25B560B / 25B124B / 5.5B124B / 5.1B
Ant Ling’s own price per million tokens, input / output (CNY)No price listed; free for the first 2 weeks after launch¥2.00 / ¥6.00 (promotion: ¥0.50 / ¥1.50)¥0.50 / ¥1.50 (promotion: ¥0.14 / ¥0.42)¥0.50 / ¥1.50 (promotion: ¥0.14 / ¥0.42)
OpenRouter id and list price per million tokens, input / outputling-3.2-flash:free, $0 / $0ling-3.1-flash, $0.045 / $0.135ling-3.0-flash-vl, $0.06 / $0.18ling-3.0-flash, $0.021 / $0.063
OpenRouter context and input1,048,576 tokens; text and image1,000,000 tokens; text131,072 tokens; text, image and video262,144 tokens; text
Public weights on Hugging FaceNone foundYes (inclusionAI/Ling-3.1-flash; licence field reads MIT)Yes (inclusionAI/Ling-3.0-flash-VL)Yes (inclusionAI/Ling-3.0-flash)

Source: developer.ant-ling.comSource: developer.ant-ling.comSource: openrouter.aiSource: huggingface.coSource: huggingface.co

Get told when it is available

Sign in and press the button: we store your account and this model, nothing else. No emails are sent yet.

Nearest available

Alternatives to Ling 3.2 Flash you can use now

Inkling 2 Small

Thinking Machines Lab

$0.30 / $1 (per million tokens)

  • Thinking Machines Lab announced Inkling 2 Small on November 5, 2026, and made it available on its Tinker platform.Source: thinkingmachines.ai · checked 2026-11-14
  • A mixture-of-experts model with 594B total parameters and 17B active, with image input and variable effort.Source: thinkingmachines.ai · checked 2026-11-14
  • Text and image input, text output; context window up to 1M tokens; each token is routed to 8 of 384 experts.Source: huggingface.co · checked 2026-11-14

Hy4 Flash

Tencent

$0.04 / $0.14 (per million tokens)

  • Tencent’s Hy4 flash announcement, dated November 16, 2026, presents it as the newest member of the Hy model family. Tencent recommends it for most everyday tasks (deep research, document processing, PowerPoint generation, simple front-end work, computer use) and points complex tasks such as long-horizon software engineering to Hy4. Tencent Cloud’s API guide calls it "a leaner, more efficient everyday workhorse".Source: tencent.com · checked 2026-11-17
  • The Tencent announcement is dated November 16, 2026, and Tencent Cloud’s model, pricing and API pages for hy4-flash were last updated the same day. The announcement says only that Hy4 flash is available now; it gives no separate release date or version name.Source: tencent.com · checked 2026-11-17
  • Tencent says Hy4 flash has 224B total parameters (including 64B Engram parameters), 12B active parameters, a context of up to 1M tokens and multimodal input; the announcement does not state a maximum output or the output types. It says the KV cache is 1/6 the size of Hy4’s and that decoding exceeds 150 tokens per second. These are the maker’s own figures, not tested by us.Source: tencent.com · checked 2026-11-17

FAQ

Ling 3.2 Flash FAQ

What is Ling 3.2 Flash?

Ling 3.2 Flash is the latest model in Ant Group’s Ling series, published through the Ant Ling platform and listed on OpenRouter under the maker name inclusionAI. It has 560B total parameters with about 25B active, reads images, text and video, and has a 1M-token context window. Ant Ling aims it at complex visual understanding, multimodal agents, 3D understanding and medical imaging.

When was Ling 3.2 Flash released?

Ant Ling’s changelog dates the launch November 5, 2026, platform version v2.0.6. OpenRouter’s page shows November 13, 2026 as its release date for the model, eight days later.

Is Ling 3.2 Flash free, and for how long?

There are two separate offers. On Ant Ling’s own platform it is free for the first 2 weeks after launch, then original prices resume; the pricing page gives no end date and no price for it. On OpenRouter the only listing is inclusionai/ling-3.2-flash:free at $0 per million input and output tokens, with per-minute and per-day request limits for free variants, and neither page says how long that listing stays. Counted from November 5, the two weeks end around November 19, 2026 (our arithmetic), so check before you rely on it.

How big is Ling 3.2 Flash and what can it read?

Ant Ling gives 560B total parameters with about 25B active, a 1M-token context and native image, text and video input. OpenRouter lists a 1,048,576-token context, up to 262,144 completion tokens and up to 200 images per request, but only text and image input for the free listing.

Are the weights of Ling 3.2 Flash open?

We found no public weights. inclusionAI’s Hugging Face list has repositories for Ling-3.0 and Ling-3.1-flash but none for Ling-3.2, and the Hugging Face id OpenRouter gives did not load. Ant Ling calls the Ling series open-sourced, but its Ling-3.2-flash section does not mention weights or a licence.

How good is Ling 3.2 Flash?

Only the maker’s own figures are available to us. AICoder reports that Ant Ling’s launch post compares it with Claude Opus 5 (max): OmniDocBench1.5 92.21 vs 90.56, ZEROBench (Sub) 43.71 vs 39.52 and Vision2Web 85.86 vs 80.20. We could not open the post and have not run these tests.

Can I chat with Ling 3.2 Flash on this site?

No. This site’s chat runs paid OpenRouter endpoints with a fixed per-token price, and OpenRouter lists Ling 3.2 Flash only as a $0 :free variant, which the chat does not run. A model with text and image input and a 1M-token context that you can try here is Inkling 2 Small: a typical reply costs about 1 credit (a credit is about $0.01 in the $5 pack), billed by its real token cost, at least 1 credit. Hy4 Flash is another low-priced option.

Is this the official Ling 3.2 Flash website?

No. This page is unofficial and not affiliated with or endorsed by inclusionAI. Facts link to the pages we read.