--- title: 'Qwen3-VL-235B-A22B-Thinking API: Model ID and Vision Input | RunAPI' description: Call qwen3-vl-235b-a22b-thinking through RunAPI's OpenAI-compatible Chat Completions API. Thinking-only vision model with 131K context and image input. url: https://runapi.ai/models/qwen/qwen3-vl-235b-a22b-thinking.md canonical: https://runapi.ai/models/qwen/qwen3-vl-235b-a22b-thinking locale: en site: RunAPI image: https://runapi.ai/assets/opengraph-bfaaa0f0.png alternates: en: https://runapi.ai/models/qwen/qwen3-vl-235b-a22b-thinking zh-CN: https://runapi.ai/zh-CN/models/qwen/qwen3-vl-235b-a22b-thinking es: https://runapi.ai/es/models/qwen/qwen3-vl-235b-a22b-thinking ja: https://runapi.ai/ja/models/qwen/qwen3-vl-235b-a22b-thinking de: https://runapi.ai/de/models/qwen/qwen3-vl-235b-a22b-thinking fr: https://runapi.ai/fr/models/qwen/qwen3-vl-235b-a22b-thinking pt-BR: https://runapi.ai/pt-BR/models/qwen/qwen3-vl-235b-a22b-thinking ko: https://runapi.ai/ko/models/qwen/qwen3-vl-235b-a22b-thinking it: https://runapi.ai/it/models/qwen/qwen3-vl-235b-a22b-thinking nl: https://runapi.ai/nl/models/qwen/qwen3-vl-235b-a22b-thinking pl: https://runapi.ai/pl/models/qwen/qwen3-vl-235b-a22b-thinking tr: https://runapi.ai/tr/models/qwen/qwen3-vl-235b-a22b-thinking zh-TW: https://runapi.ai/zh-TW/models/qwen/qwen3-vl-235b-a22b-thinking zh-HK: https://runapi.ai/zh-HK/models/qwen/qwen3-vl-235b-a22b-thinking ar: https://runapi.ai/ar/models/qwen/qwen3-vl-235b-a22b-thinking x-default: https://runapi.ai/models/qwen/qwen3-vl-235b-a22b-thinking model: Qwen provider: Alibaba variant_id: qwen3-vl-235b-a22b-thinking sdk_model_id: qwen3-vl-235b-a22b-thinking modality: text billing_unit: 1K tokens status: retired --- > HTML version: https://runapi.ai/models/qwen/qwen3-vl-235b-a22b-thinking > Site index for agents: https://runapi.ai/llms.txt # qwen3-vl-235b-a22b-thinking API **Retired:** Retired on 2026-10-10. Call qwen3-vl-235b-a22b-thinking from the Qwen family through RunAPI. Billing is per token with no subscription, and a request that fails before its response starts is not charged. **Alibaba / [Qwen](https://runapi.ai/models/qwen.md)** **Retired · text · Commercial OK** ## What Qwen3-VL-235B-A22B-Thinking is built for Qwen3-VL-235B-A22B-Thinking is the reasoning version of Qwen3-VL, the flagship vision-language model family from Alibaba Qwen. It is a Mixture-of-Experts model with 235B total parameters and 22B active per token, released with open weights under Apache 2.0 on September 23, 2025. Alibaba tunes it for multimodal reasoning, with a focus on STEM and math problems, OCR, and image and video understanding. The model is thinking-only: every response starts with a reasoning step, and the step cannot be switched off. Alibaba's API serves it with a 131,072-token context window and up to 32,768 output tokens. On RunAPI, call qwen3-vl-235b-a22b-thinking through the OpenAI-compatible Chat Completions endpoint (/v1/chat/completions) with an OpenAI SDK pointed at https://runapi.ai/v1. Send images as image_url parts next to the text in messages[].content[]. Alibaba now lists Qwen3-VL among its previous-generation models, because the Qwen3.5 and Qwen3.6 models read images natively. **Model ID:** qwen3-vl-235b-a22b-thinking **Released:** September 23, 2025 **Architecture:** MoE, 235B total / 22B active, Apache 2.0 weights **Context window:** 131,072 tokens **Max output:** 32,768 tokens **Thinking:** Always on **Endpoint on RunAPI:** OpenAI Chat Completions, text and image input ### Switch variant - [qwen3-coder-next](https://runapi.ai/models/qwen/qwen3-coder-next.md) - [qwen3-vl-235b-a22b-instruct](https://runapi.ai/models/qwen/qwen3-vl-235b-a22b-instruct.md) - **qwen3-vl-235b-a22b-thinking** (current) - [qwen3.5-122b-a10b](https://runapi.ai/models/qwen/qwen3.5-122b-a10b.md) - [qwen3.5-27b](https://runapi.ai/models/qwen/qwen3.5-27b.md) - [qwen3.5-35b-a3b](https://runapi.ai/models/qwen/qwen3.5-35b-a3b.md) - [qwen3.5-397b-a17b](https://runapi.ai/models/qwen/qwen3.5-397b-a17b.md) - [qwen3.6-27b](https://runapi.ai/models/qwen/qwen3.6-27b.md) - [qwen3.6-35b-a3b](https://runapi.ai/models/qwen/qwen3.6-35b-a3b.md) 9 versions available qwen3-vl-235b-a22b-thinking is one of the Qwen models on RunAPI. It uses the same API key as every other model, and you switch to it by changing the model ID in your request. Each call is synchronous and billed in USD by input, output, and cache tokens, with no subscription or minimum spend. A request that fails before the response starts is not charged; once a response is streaming, the tokens already generated are billed, even if the client disconnects. - Per-token billing in USD with no subscription - No charge when a request fails before its response starts - Token-by-token streaming - Switch models by changing one model ID - One API key and one bill for every model and provider ## Technical details | Field | Value | |---|---| | Provider | Alibaba | | Model ID | qwen3-vl-235b-a22b-thinking | | Modality | Text | | Task types | Synchronous | | API endpoint | `/v1/chat/completions` | | | `/v1/responses` | | | `/v1/messages` | | | `/v1beta/models/qwen3-vl-235b-a22b-thinking:generateContent` | | | `/v1beta/models/qwen3-vl-235b-a22b-thinking:streamGenerateContent` | | Billing unit | 1K tokens | | Catalog status | Retired | **HOW IT WORKS** ## Use This Model in 3 Steps 1. **Get Your Key** - Sign up for free and create an API key for qwen3-vl-235b-a22b-thinking from the dashboard. 2. **Send a Request** - POST to /v1/chat/completions with the qwen3-vl-235b-a22b-thinking model slug and your parameters. 3. **Get Results** - Read the completed qwen3-vl-235b-a22b-thinking response directly from the endpoint. ## What this model offers - **Endpoint coverage** — Exposes 1 public endpoint(s): chat_completion. - **Usage-based billing** — Metered by 1K tokens with no subscription requirement. **API REFERENCE** ## Endpoint Specification | Field | Value | |---|---| | Endpoint | `POST /v1/chat/completions` | | Model ID | `qwen3-vl-235b-a22b-thinking` | | Task Types | `chat_completion` | ## Request parameters No request parameters are published for this endpoint. **Request example** ```json { "model": "qwen3-vl-235b-a22b-thinking", "messages": [ { "role": "user", "content": "Read this screenshot of a failing dashboard, explain what the error message means, and suggest the code change that would fix it." } ] } ``` ### Model skill — qwen3-vl-235b-a22b-thinking Install the skill once, then use the variant ID from this page while building. Base URL: `https://runapi.ai` | Endpoint | Protocol | |---|---| | `POST /v1/chat/completions` | OpenAI compatible | | `POST /v1/responses` | OpenAI Responses | | `POST /v1/messages` | Anthropic compatible | | `POST /v1beta/models/qwen3-vl-235b-a22b-thinking:generateContent` | Gemini generateContent | | `POST /v1beta/models/qwen3-vl-235b-a22b-thinking:streamGenerateContent` | Gemini streamGenerateContent | Use the OpenAI / Anthropic / Google Gen AI SDK with your RunAPI API key and `model: "qwen3-vl-235b-a22b-thinking"`. ### Use qwen3-vl-235b-a22b-thinking with a model skill 1. **Install** — Install the model skill for the Qwen line. It loads docs, request formats, and pricing notes into your coding workspace. 2. **Configure** — Point your SDK at https://runapi.ai and set your RunAPI API key. The same key works for every model and provider. 3. **Call** — Send a request in an API format this model supports, with the model field set to the model ID on this page. RunAPI forwards it to the provider. 4. **Receive** — Read the response from the same call, or stream it as it is generated. The usage field shows the tokens you are billed for. ### What's different about qwen3-vl-235b-a22b-thinking - **VS QWEN3-CODER-NEXT: Thinking-only Qwen3-VL vision MoE; 131K context; image input via OpenAI-compatible API** — Open-weight Qwen coding model for agents (3B active); 256K context; OpenAI-compatible API - **VS QWEN3-VL-235B-A22B-INSTRUCT: Thinking-only Qwen3-VL vision MoE; 131K context; image input via OpenAI-compatible API** — Qwen3-VL vision MoE with direct answers; 131K context; image input via OpenAI-compatible API - **VS QWEN3.5-122B-A10B: Thinking-only Qwen3-VL vision MoE; 131K context; image input via OpenAI-compatible API** — Open-weight Qwen3.5 vision MoE (10B active); 256K context; OpenAI-compatible API ## USE CASES **Customer support** Answer customer questions from a private knowledge base, reducing ticket volume. **Document analysis** Draft contract summaries and flag key clauses for attorney review. **Code generation** Auto-generate unit tests, code reviews, and refactoring suggestions in CI. ## Frequently asked questions about Qwen3-VL-235B-A22B-Thinking ### What is the difference between the Thinking and Instruct versions of Qwen3-VL-235B-A22B? Both use the same 235B Mixture-of-Experts model with 22B active parameters. The Thinking version always reasons before it answers and is tuned for STEM, math, and multimodal reasoning. The Instruct version answers directly and is aimed at visual coding, spatial perception, long-video understanding, and OCR. ### How do I send an image to qwen3-vl-235b-a22b-thinking? Use the Chat Completions endpoint and add an image_url part next to the text part in messages[].content[], the same request shape the OpenAI SDK uses for vision. Alibaba also lists video input for this model; video requests are not verified on RunAPI. ### Can I turn thinking off on this model? No. Alibaba ships it as a thinking-only model. If you want answers without a reasoning step, call qwen3-vl-235b-a22b-instruct instead. ### Should I use Qwen3-VL or a Qwen3.5 model for image tasks? Alibaba lists Qwen3-VL among its previous-generation models and says the vision capability of Qwen3.5-122B-A10B exceeds Qwen3-VL-235B. For new image workloads, try a Qwen3.5 or Qwen3.6 model first, and keep Qwen3-VL for prompts and evaluations already built on it. ### How do I select qwen3-vl-235b-a22b-thinking? Pass the model ID shown in the quickstart. ### What's the rate limit on this variant? Per-key rate limits scale with your usage tier. The pricing page shows current limits. If you need higher throughput, contact support to discuss tier upgrades. ### Can I switch variants later? Yes. Variant is a parameter in the request. Switch by changing the model ID — no code changes, no re-authentication, no separate billing setup. All variants share the same API key and request shape. ### Does it stream? Where streaming is available, RunAPI streams end-to-end. LLM models support token-level streaming. Media models use async task polling or webhook callbacks for result delivery. ### Where do I report quality issues? Open an issue on the public GitHub repo or email support at contact@runapi.ai. Include the task ID and model ID so the team can investigate the specific generation. ### Do I need a separate provider account? No. Your RunAPI API key is enough to access this variant and every other model in the catalog. You do not need accounts with the underlying provider. ### RELATED - [qwen3.5-35b-a3b](https://runapi.ai/models/qwen/qwen3.5-35b-a3b.md) — by Alibaba · Input $0.25 / 1M tokens | Output $2.00 / 1M tokens - [qwen3.5-27b](https://runapi.ai/models/qwen/qwen3.5-27b.md) — by Alibaba · Input $0.30 / 1M tokens | Output $2.40 / 1M tokens - [qwen3.6-35b-a3b](https://runapi.ai/models/qwen/qwen3.6-35b-a3b.md) — by Alibaba · Input $0.38 / 1M tokens | Output $2.25 / 1M tokens - [qwen3.5-122b-a10b](https://runapi.ai/models/qwen/qwen3.5-122b-a10b.md) — by Alibaba · Input $0.40 / 1M tokens | Output $3.20 / 1M tokens ### Start building with Qwen. - [Create free account](https://runapi.ai/login) - [Read the quickstart →](https://runapi.ai/docs/api/openai/chat-completions) ## Building with a team? We're here to help with enterprise setup, integrations, and technical questions. [Contact Us](https://runapi.ai/contact.md) --- ## More from RunAPI - [Home](https://runapi.ai/.md) ### Product - [Model Catalog](https://runapi.ai/models.md) - [Pricing](https://runapi.ai/pricing.md) - [Providers](https://runapi.ai/models#all-models) ### Developers - [Documentation](https://runapi.ai/docs/guides) - [SDKs](https://runapi.ai/sdk.md) - [CLI](https://runapi.ai/cli.md) - [MCP Server](https://runapi.ai/mcp.md) - [x402](https://runapi.ai/x402.md) ### Guides - [Claude Code vs Cursor](https://runapi.ai/claude-code-vs-cursor.md) - [Cursor API Setup](https://runapi.ai/cursor-api-setup.md) - [RunAPI vs OpenRouter](https://runapi.ai/openrouter-alternative.md) ### Company - [Enterprise](https://runapi.ai/contact.md) - [Contact](https://runapi.ai/contact.md) ### Legal - [Terms](https://runapi.ai/terms.md) - [Privacy](https://runapi.ai/privacy.md) - [Site index for agents](https://runapi.ai/llms.txt) Contact: contact@runapi.ai ## Structured data ```json [ { "@context": "https://schema.org", "inLanguage": "en", "@type": "WebSite", "name": "RunAPI", "url": "https://runapi.ai/", "potentialAction": { "@type": "SearchAction", "target": { "@type": "EntryPoint", "urlTemplate": "https://runapi.ai/models?q={search_term_string}" }, "query-input": "required name=search_term_string" } }, { "@context": "https://schema.org", "inLanguage": "en", "@type": "Organization", "name": "RunAPI", "url": "https://runapi.ai/", "logo": { "@type": "ImageObject", "url": "https://runapi.ai/icon.svg" }, "sameAs": [ "https://github.com/runapi-ai" ] }, { "@context": "https://schema.org", "inLanguage": "en", "@type": "SoftwareApplication", "name": "Qwen qwen3-vl-235b-a22b-thinking", "applicationCategory": "DeveloperApplication" }, { "@context": "https://schema.org", "inLanguage": "en", "@type": "BreadcrumbList", "itemListElement": [ { "@type": "ListItem", "position": 1, "name": "Home", "item": "https://runapi.ai/" }, { "@type": "ListItem", "position": 2, "name": "Models", "item": "https://runapi.ai/models" }, { "@type": "ListItem", "position": 3, "name": "Qwen", "item": "https://runapi.ai/models/qwen" }, { "@type": "ListItem", "position": 4, "name": "qwen3-vl-235b-a22b-thinking", "item": "https://runapi.ai/models/qwen/qwen3-vl-235b-a22b-thinking" } ] }, { "@context": "https://schema.org", "inLanguage": "en", "@type": "FAQPage", "mainEntity": [ { "@type": "Question", "name": "What is the difference between the Thinking and Instruct versions of Qwen3-VL-235B-A22B?", "acceptedAnswer": { "@type": "Answer", "text": "Both use the same 235B Mixture-of-Experts model with 22B active parameters. The Thinking version always reasons before it answers and is tuned for STEM, math, and multimodal reasoning. The Instruct version answers directly and is aimed at visual coding, spatial perception, long-video understanding, and OCR." } }, { "@type": "Question", "name": "How do I send an image to qwen3-vl-235b-a22b-thinking?", "acceptedAnswer": { "@type": "Answer", "text": "Use the Chat Completions endpoint and add an image_url part next to the text part in messages[].content[], the same request shape the OpenAI SDK uses for vision. Alibaba also lists video input for this model; video requests are not verified on RunAPI." } }, { "@type": "Question", "name": "Can I turn thinking off on this model?", "acceptedAnswer": { "@type": "Answer", "text": "No. Alibaba ships it as a thinking-only model. If you want answers without a reasoning step, call qwen3-vl-235b-a22b-instruct instead." } }, { "@type": "Question", "name": "Should I use Qwen3-VL or a Qwen3.5 model for image tasks?", "acceptedAnswer": { "@type": "Answer", "text": "Alibaba lists Qwen3-VL among its previous-generation models and says the vision capability of Qwen3.5-122B-A10B exceeds Qwen3-VL-235B. For new image workloads, try a Qwen3.5 or Qwen3.6 model first, and keep Qwen3-VL for prompts and evaluations already built on it." } }, { "@type": "Question", "name": "How do I select qwen3-vl-235b-a22b-thinking?", "acceptedAnswer": { "@type": "Answer", "text": "Pass the model ID shown in the quickstart." } }, { "@type": "Question", "name": "What's the rate limit on this variant?", "acceptedAnswer": { "@type": "Answer", "text": "Per-key rate limits scale with your usage tier. The pricing page shows current limits. If you need higher throughput, contact support to discuss tier upgrades." } }, { "@type": "Question", "name": "Can I switch variants later?", "acceptedAnswer": { "@type": "Answer", "text": "Yes. Variant is a parameter in the request. Switch by changing the model ID — no code changes, no re-authentication, no separate billing setup. All variants share the same API key and request shape." } }, { "@type": "Question", "name": "Does it stream?", "acceptedAnswer": { "@type": "Answer", "text": "Where streaming is available, RunAPI streams end-to-end. LLM models support token-level streaming. Media models use async task polling or webhook callbacks for result delivery." } }, { "@type": "Question", "name": "Where do I report quality issues?", "acceptedAnswer": { "@type": "Answer", "text": "Open an issue on the public GitHub repo or email support at contact@runapi.ai. Include the task ID and model ID so the team can investigate the specific generation." } }, { "@type": "Question", "name": "Do I need a separate provider account?", "acceptedAnswer": { "@type": "Answer", "text": "No. Your RunAPI API key is enough to access this variant and every other model in the catalog. You do not need accounts with the underlying provider." } } ] } ] ```