Source: https://ailoot.ai/models/en/reflection-beam-1-0-501b-a23b/ Canonical: https://ailoot.ai/models/en/reflection-beam-1-0-501b-a23b/ Language: en # Reflection: Beam 1.0 501B A23B [Markdown](https://ailoot.ai/models/en/reflection-beam-1-0-501b-a23b/index.md) · [JSON](https://ailoot.ai/models/en/reflection-beam-1-0-501b-a23b/facts.json) ## [Reflection: Beam 1.0 501B A23B](https://ailoot.ai/models/en/reflection-beam-1-0-501b-a23b/) `reflection/beam-1.0-501b-a23b` ReflectionTextInput $0.0005 / 1KOutput $0.002 / 1K Reflection’s large-context text MoE model for software engineering, mathematical reasoning and step-by-step problem analysis. ### Beam 1.0 from Reflection Beam 1.0 501B A23B is a Reflection text model for programming, reasoning and tasks divided into several connected steps. Its sparse MoE architecture has 501.3 billion total parameters and approximately 23.4 billion active parameters per token. The model was developed primarily for English; consider this when choosing the language of source code, technical material and detailed instructions. The architecture combines local and global attention, and the model’s preparation includes source code, technical documentation, scientific material and educational content. A separate training stage developed long-context capabilities using structured repositories, extended tasks and large documents. During reinforcement learning, the developer combined programming, terminal-work and STEM tasks. ### The model’s intended tasks Reflection highlights software engineering, mathematical and scientific reasoning, and sequential problem-solving. These areas lend themselves to substantive text conversations: explaining relationships between program components, investigating a logic error, proposing an algorithm change, planning a hypothesis test or summarizing technical material. Beam places particular emphasis on efficient use of reasoning tokens. Training rewarded successful solutions that reduced unnecessary tokens. This describes the developer’s approach; you should still specify the desired response length in your request—for example, a brief conclusion, a detailed proof or a patch with an explanation. ### A large text context in Ailoot API The connection has a context window of 1,048,576 tokens, shared by supplied material, instructions, history and the response being generated. The model’s metadata lists output of up to 131,072 tokens. This capacity lets you provide several connected texts: a specification, implementation fragments, verification results and questions about them. Give the material names and ask the model to identify the passage supporting each conclusion, making extensive analyses easier to check. For a coding task, for example, first ask for a dependency map based on the supplied files, then identify the location of a change and only then prepare the implementation. For a scientific task, specify the initial conditions, notation and required level of rigor. These request structures make it possible to compare a useful result with the source material. The connection accepts text and returns text through Messages, Responses, Chat Completions and text-based generateContent. Code and recommendations arrive as the model’s response; command execution and access to external data depend on your application’s capabilities. ### Available through Ailoot API Input types `text` Output types `text` Context Not specified in the catalogue Max. output Not specified in the catalogue Input $0.0005 */ 1K* Output $0.002 */ 1K* Cache read $0.0001 */ 1K* Input types `text` Output types `text` Context Not specified in the catalogue Max. output Not specified in the catalogue Input $0.0005 */ 1K* Output $0.002 */ 1K* Cache read $0.0001 */ 1K* Input types `text` Output types `text` Context Not specified in the catalogue Max. output Not specified in the catalogue Input $0.0005 */ 1K* Output $0.002 */ 1K* Cache read $0.0001 */ 1K* Input types `text` Output types `text` Context Not specified in the catalogue Max. output Not specified in the catalogue Input $0.0005 */ 1K* Output $0.002 */ 1K* Cache read $0.0001 */ 1K* `reflection/beam-1.0-501b-a23b` · JSON ``` {"about":"### Beam 1.0 from Reflection\n\nBeam 1.0 501B A23B is a Reflection text model for programming, reasoning and tasks divided into several connected steps. Its sparse MoE architecture has 501.3 billion total parameters and approximately 23.4 billion active parameters per token. The model was developed primarily for English; consider this when choosing the language of source code, technical material and detailed instructions.\n\nThe architecture combines local and global attention, and the model’s preparation includes source code, technical documentation, scientific material and educational content. A separate training stage developed long-context capabilities using structured repositories, extended tasks and large documents. During reinforcement learning, the developer combined programming, terminal-work and STEM tasks.\n\n### The model’s intended tasks\n\nReflection highlights software engineering, mathematical and scientific reasoning, and sequential problem-solving. These areas lend themselves to substantive text conversations: explaining relationships between program components, investigating a logic error, proposing an algorithm change, planning a hypothesis test or summarizing technical material.\n\nBeam places particular emphasis on efficient use of reasoning tokens. Training rewarded successful solutions that reduced unnecessary tokens. This describes the developer’s approach; you should still specify the desired response length in your request—for example, a brief conclusion, a detailed proof or a patch with an explanation.\n\n### A large text context in Ailoot API\n\nThe connection has a context window of 1,048,576 tokens, shared by supplied material, instructions, history and the response being generated. The model’s metadata lists output of up to 131,072 tokens. This capacity lets you provide several connected texts: a specification, implementation fragments, verification results and questions about them. Give the material names and ask the model to identify the passage supporting each conclusion, making extensive analyses easier to check.\n\nFor a coding task, for example, first ask for a dependency map based on the supplied files, then identify the location of a change and only then prepare the implementation. For a scientific task, specify the initial conditions, notation and required level of rigor. These request structures make it possible to compare a useful result with the source material.\n\nThe connection accepts text and returns text through Messages, Responses, Chat Completions and text-based generateContent. Code and recommendations arrive as the model’s response; command execution and access to external data depend on your application’s capabilities.","alias_of":null,"aliases":[],"author":"Reflection","canonical_model":"reflection/beam-1.0-501b-a23b","ctx":null,"formats":["chat","gemini","messages","responses"],"id":"reflection/beam-1.0-501b-a23b","input":["text"],"kind":"text","max_out":null,"methods":[{"cache_read":0.1,"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":0.5,"input":["text"],"label":"Anthropic Messages","max_out":null,"max_tool_calls":null,"out":2.0,"output":["text"],"path":"/v1/messages","request_model_id":"reflection/beam-1.0-501b-a23b","route_label":"Ultra","status_reads":[],"tiers":[],"tool_prices":null,"unit":"token","units":[]},{"cache_read":0.1,"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":0.5,"input":["text"],"label":"OpenAI Responses","max_out":null,"max_tool_calls":null,"out":2.0,"output":["text"],"path":"/v1/responses","request_model_id":"reflection/beam-1.0-501b-a23b","route_label":"Ultra","status_reads":[],"tiers":[],"tool_prices":null,"unit":"token","units":[]},{"cache_read":0.1,"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":0.5,"input":["text"],"label":"OpenAI Chat","max_out":null,"max_tool_calls":null,"out":2.0,"output":["text"],"path":"/v1/chat/completions","request_model_id":"reflection/beam-1.0-501b-a23b","route_label":"Ultra","status_reads":[],"tiers":[],"tool_prices":null,"unit":"token","units":[]},{"cache_read":0.1,"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":0.5,"input":["text"],"label":"Gemini generateContent","max_out":null,"max_tool_calls":null,"out":2.0,"output":["text"],"path":"/v1beta/models/reflection/beam-1.0-501b-a23b:generateContent","request_model_id":"reflection/beam-1.0-501b-a23b","route_label":"Ultra","status_reads":[],"tiers":[],"tool_prices":null,"unit":"token","units":[]}],"model_aliases":[{"canonical_model_id":"reflection/beam-1.0-501b-a23b","request_model_id":"reflection/beam-1.0-501b-a23b","route_label":"Ultra"}],"name":"Reflection: Beam 1.0 501B A23B","price":{"in":0.5,"in_label":null,"out":2.0,"out_label":null,"sort_in":0.5,"sort_out":2.0,"varies":false},"public_id":"reflection/beam-1.0-501b-a23b","short":"Reflection’s large-context text MoE model for software engineering, mathematical reasoning and step-by-step problem analysis."} ``` ## 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.