Source: https://ailoot.ai/models/en/liquid-lfm-2-5-2-6b/ Canonical: https://ailoot.ai/models/en/liquid-lfm-2-5-2-6b/ Language: en # LiquidAI: LFM2.5-2.6B [Markdown](https://ailoot.ai/models/en/liquid-lfm-2-5-2-6b/index.md) · [JSON](https://ailoot.ai/models/en/liquid-lfm-2-5-2-6b/facts.json) ## [LiquidAI: LFM2.5-2.6B](https://ailoot.ai/models/en/liquid-lfm-2-5-2-6b/) `liquid/lfm-2.5-2.6b` LiquidTextInput $0.0007 / 1KOutput $0.001 / 1K A compact Liquid AI text model for data extraction, document-based answers and sequential tasks with a large context. ### A compact model with agentic training LFM2.5-2.6B is a Liquid AI text model built for tasks involving instruction following, document processing and the sequential completion of several steps. Within the LFM2.5 family, this version received dedicated training for agentic workflows: the developer trained it on interactions with tools and typical agent working environments. That training concerns the model’s behavior; the integration determines the specific actions available in an application. LFM2.5-2.6B uses the hybrid LFM2 architecture with 2.69 billion parameters. It combines short-convolution blocks and grouped attention. Liquid AI is developing the family with on-device operation in mind, but in Ailoot API the model is accessed through a text interface: the user submits messages and receives a text response. ### Where it is useful The model suits information extraction from text, questions about supplied material and processing long instructions. For example, provide customer correspondence, ask it to consolidate the requirements into one list, then identify contradictions and draft a clarification email. For knowledge-base answers, include the retrieved passages and specify that it should rely on them, preserve names and indicate when a necessary fact is absent. Liquid AI’s documentation lists data extraction, RAG and long-context workflows among its use cases. The developer explicitly limits the recommended scope: agentic programming and tasks requiring substantial factual knowledge without supplied sources are not recommended for this version. When working with industry terminology, include definitions and source documents in the request. ### Context and languages This model has a stated context of 128,000 tokens in Ailoot API. It includes the task material and conversation history; plan the response size alongside them. The developer’s model card lists 16 languages, including Russian, English, Chinese, German and Spanish. This supports multilingual document and correspondence processing, with the desired output language and format specified in the instructions. ### Available through Ailoot API Input types `text` Output types `text` Context Not specified in the catalogue Max. output Not specified in the catalogue Input $0.0007 */ 1K* Output $0.001 */ 1K* Cache read $0.00008 */ 1K* Input types `text` Output types `text` Context Not specified in the catalogue Max. output Not specified in the catalogue Input $0.0007 */ 1K* Output $0.001 */ 1K* Cache read $0.00008 */ 1K* Input types `text` Output types `text` Context Not specified in the catalogue Max. output Not specified in the catalogue Input $0.0007 */ 1K* Output $0.001 */ 1K* Cache read $0.00008 */ 1K* Input types `text` Output types `text` Context Not specified in the catalogue Max. output Not specified in the catalogue Input $0.0007 */ 1K* Output $0.001 */ 1K* Cache read $0.00008 */ 1K* `liquid/lfm-2.5-2.6b` · JSON ``` {"about":"### A compact model with agentic training\n\nLFM2.5-2.6B is a Liquid AI text model built for tasks involving instruction following, document processing and the sequential completion of several steps. Within the LFM2.5 family, this version received dedicated training for agentic workflows: the developer trained it on interactions with tools and typical agent working environments. That training concerns the model’s behavior; the integration determines the specific actions available in an application.\n\nLFM2.5-2.6B uses the hybrid LFM2 architecture with 2.69 billion parameters. It combines short-convolution blocks and grouped attention. Liquid AI is developing the family with on-device operation in mind, but in Ailoot API the model is accessed through a text interface: the user submits messages and receives a text response.\n\n### Where it is useful\n\nThe model suits information extraction from text, questions about supplied material and processing long instructions. For example, provide customer correspondence, ask it to consolidate the requirements into one list, then identify contradictions and draft a clarification email. For knowledge-base answers, include the retrieved passages and specify that it should rely on them, preserve names and indicate when a necessary fact is absent.\n\nLiquid AI’s documentation lists data extraction, RAG and long-context workflows among its use cases. The developer explicitly limits the recommended scope: agentic programming and tasks requiring substantial factual knowledge without supplied sources are not recommended for this version. When working with industry terminology, include definitions and source documents in the request.\n\n### Context and languages\n\nThis model has a stated context of 128,000 tokens in Ailoot API. It includes the task material and conversation history; plan the response size alongside them. The developer’s model card lists 16 languages, including Russian, English, Chinese, German and Spanish. This supports multilingual document and correspondence processing, with the desired output language and format specified in the instructions.","alias_of":null,"aliases":[],"author":"Liquid","canonical_model":"liquid/lfm-2.5-2.6b","ctx":null,"formats":["chat","gemini","messages","responses"],"id":"liquid/lfm-2.5-2.6b","input":["text"],"kind":"text","max_out":null,"methods":[{"cache_read":0.08,"cache_write":null,"cache_write_1h":null,"cache_write_5m":null,"conditional_meter":null,"ctx":null,"family":"anthropic","fast_multiplier":1,"flex_multiplier":1,"fmt":"messages","http":"POST","img_in":null,"img_out":null,"in":0.7,"input":["text"],"label":"Anthropic Messages","max_out":null,"max_tool_calls":null,"out":1.0,"output":["text"],"path":"/v1/messages","request_model_id":"liquid/lfm-2.5-2.6b","route_label":"Ultra","status_reads":[],"tiers":[],"tool_prices":null,"unit":"token","units":[]},{"cache_read":0.08,"cache_write":null,"cache_write_1h":null,"cache_write_5m":null,"conditional_meter":null,"ctx":null,"family":"openai","fast_multiplier":1,"flex_multiplier":1,"fmt":"responses","http":"POST","img_in":null,"img_out":null,"in":0.7,"input":["text"],"label":"OpenAI Responses","max_out":null,"max_tool_calls":null,"out":1.0,"output":["text"],"path":"/v1/responses","request_model_id":"liquid/lfm-2.5-2.6b","route_label":"Ultra","status_reads":[],"tiers":[],"tool_prices":null,"unit":"token","units":[]},{"cache_read":0.08,"cache_write":null,"cache_write_1h":null,"cache_write_5m":null,"conditional_meter":null,"ctx":null,"family":"openai","fast_multiplier":1,"flex_multiplier":1,"fmt":"chat","http":"POST","img_in":null,"img_out":null,"in":0.7,"input":["text"],"label":"OpenAI Chat","max_out":null,"max_tool_calls":null,"out":1.0,"output":["text"],"path":"/v1/chat/completions","request_model_id":"liquid/lfm-2.5-2.6b","route_label":"Ultra","status_reads":[],"tiers":[],"tool_prices":null,"unit":"token","units":[]},{"cache_read":0.08,"cache_write":null,"cache_write_1h":null,"cache_write_5m":null,"conditional_meter":null,"ctx":null,"family":"gemini","fast_multiplier":1,"flex_multiplier":1,"fmt":"gemini","http":"POST","img_in":null,"img_out":null,"in":0.7,"input":["text"],"label":"Gemini generateContent","max_out":null,"max_tool_calls":null,"out":1.0,"output":["text"],"path":"/v1beta/models/liquid/lfm-2.5-2.6b:generateContent","request_model_id":"liquid/lfm-2.5-2.6b","route_label":"Ultra","status_reads":[],"tiers":[],"tool_prices":null,"unit":"token","units":[]}],"model_aliases":[{"canonical_model_id":"liquid/lfm-2.5-2.6b","request_model_id":"liquid/lfm-2.5-2.6b","route_label":"Ultra"}],"name":"LiquidAI: LFM2.5-2.6B","price":{"in":0.7,"in_label":null,"out":1.0,"out_label":null,"sort_in":0.7,"sort_out":1.0,"varies":false},"public_id":"liquid/lfm-2.5-2.6b","short":"A compact Liquid AI text model for data extraction, document-based answers and sequential tasks with a large context."} ``` ## Questions How do I choose a model for my task? First identify the required input and output: text, image, audio or video. Then compare compatible API methods, the exact model ID or current alias target, price and billing unit. GET /v1/models is free and lists models available to your key. Does each model support every API method and media type? Use the API methods and input/output types listed on the model card. Support for text, images, audio or video depends on the selected model and method; check their request example before sending your data. What is the difference between a model ID and an alias? An exact model ID names a specific model. An alias points to a model and its target can change. Check the current alias target in the catalogue; use the exact ID when your integration needs a specific model. How do I compare prices before sending a request? Check the model card for the price and billing unit of the selected API method. Compare the same unit and input/output type; request history shows actual usage and the charged amount. What does a subscription give me? A discount on token purchases for regular use. Choose a plan for your request volume; subscription terms and pricing are shown at checkout. How do I apply a promo code for up to 50% off? Enter the promo code when buying tokens. The total updates according to the promotion terms; you can see the discount before paying. ## Ailoot · The company and its project portfolio Ailoot combines established operating experience, a large client base, stable service and ongoing KPI control. The company and its project portfolio: millions of customers; monthly sales in the billions. Ailoot brings that operating experience to API access for AI features in customer products, agents, applications and automations.