Tasks and inputs
LongCat 2.5 Preview is listed for Image-text-to-text, Text generation, Coding & Code Generation. It accepts text, image and produces text. Use the publisher’s input format and preprocessing for your chosen deployment.
SyncDev model directory · Profile updated Oct 8, 2026
LongCat 2.5 Preview is Meituan’s hosted multimodal reasoning model for coding and agent work, with a one-million-token context.
The specification score summarizes recorded features and limits. It is not a benchmark result.
Recorded modality support shows what this model can accept and produce. An undocumented modality is shown as unknown rather than assumed to be unsupported.
Accepts
Produces
LongCat 2.5 Preview is Meituan’s hosted multimodal reasoning model for coding and agent work, with a one-million-token context.
LongCat 2.5 Preview adds image understanding to the publisher’s coding and agent workflow offering. The direct API supports OpenAI-compatible Chat Completions and Anthropic-compatible Messages. The documented context is one million tokens, with up to 131,072 output tokens. Use LongCat-2.5-Preview as the model identifier.
The Chat Completions interface exposes a thinking object with enabled and disabled modes and supports streaming responses. Configure the client against the documented LongCat endpoint and check which format your coding tool expects. The simple text example is not a complete image request: follow the publisher’s multimodal instructions before sending an image.
The rate card lists promotional prices of $0.30 per million uncached input tokens, $0.006 for cached input and $1.20 for output. These are temporary discounted rates. The billing record for your actual account remains the reference for a production budget, including any reasoning and retries.
Give the assistant a small repository issue that requires inspecting a screenshot and modifying code. Review the final patch against the visible requirement and run the same acceptance check used for other models. Then try a long document question that depends on information near the beginning and end. Record missed constraints and repeated work, as well as response time.
Model documentation · Model documentation · Model documentation · Model documentation
Compare the model on the work you need to complete.
LongCat 2.5 Preview is listed for Image-text-to-text, Text generation, Coding & Code Generation. It accepts text, image and produces text. Use the publisher’s input format and preprocessing for your chosen deployment.
Documented limits: 1M context and 131.07K maximum output. The listed route charges $0.30 input and $1.20 output per million tokens. Measure a completed task using the prompts, documents and tools your application will use. Include reasoning and retries in the result.
Use the model identifier and account requirements for the selected API route. Confirm its request limits and output format with one complete task before connecting a larger workflow.
These records share a directory category. Check the source card, output type and deployment requirements before treating a model as a replacement.
Mistral AI
Leanstral 1.5 119B A6B is an Apache-licensed model for Lean 4 formal proofs and code workflows.
Mistral AI
Leanstral 2603 is an Apache-licensed model for Lean 4 formal proofs and code workflows.
inclusionAI
Ling 2.6 1T is a trillion-parameter MIT-licensed checkpoint for complex instruction following, coding and long-document workflows.
inclusionAI
Ling 2.6 Flash is a hybrid-attention instruction checkpoint designed for concise responses and repeated agent tool workflows.
inclusionAI
Ling 3.0 Tiny is a compact MIT-licensed reasoning checkpoint with 7.9B total parameters and 1.3B active per token.
Meituan LongCat
LongCat Flash Lite Sparse adds sparse attention and native long-context support to the non-thinking LongCat Lite family.