Beam (Reflection)

Beam is a text-only large language model announced by Reflection AI on October 5, 2026 for coding, reasoning and agentic tasks.^[1]

Preview and planned weights

On October 7, 2026, Beam was a selective preview. Reflection planned to release its weights, technical report and model card later in October, with Apache 2.0 licensing for the weights.^[1]

Published specifications and beta limits

The developer documentation warns that the context window may change during beta.^[2]

Training and expert routing

Reflection reports a four-week reinforcement-learning run using 10,500 NVIDIA GB300 GPUs and more than 100 million rollouts.^[1]

Beam's expert balancing builds on auxiliary-loss-free balancing, with cosine-decayed expert-bias updates.^[1]

The preceding DeepSeek-V3 method adds an adjustable bias to each expert's routing score. Recent loads determine bias updates; the bias changes expert selection, rather than the value that weights an expert's output.^[8]

API use

The Reflection API is in beta, with access opening through a waitlist. Its OpenAI-compatible base URL is https://api.reflection.ai/openai/v1, supporting Chat Completions and Models. This is Reflection's service, rather than an OpenAI-hosted model.^[4]

Reflection documents Beam configurations for Mirror CLI, Pi, OpenCode and Hermes. Mirror CLI uses Beam by default; the other integrations use its model identifier, the compatible endpoint and a Reflection API key.^[5]

Reasoning controls

Beam-501B-A23B always reasons. Its reasoning_effort values are low, medium, high, xhigh and max; omitting the setting selects medium. Reasoning cannot be disabled through this setting.^[3]

The response separates message.reasoning_content from the final answer in message.content. Reasoning tokens count toward the completion budget. If that budget runs out during reasoning, finish_reason can be length and the answer can be null.^[3]

Tool calls and structured output

With function calling, Beam requests application-defined functions; the application executes them and returns results. Its generated JSON arguments require validation before use. The conversation must retain the assistant's tool-call message, including reasoning, followed by one tool-result message for each call.^[6]

Structured output uses response_format: json_object requests a JSON object, while json_schema supplies a schema. Strict schema mode requires every object to set additionalProperties: false and require all its properties. Applications must check for refusals and truncated output before parsing the answer.^[7]

Vendor-reported evaluation

Reflection estimates generation compute as 2 × active parameters × mean generated tokens per attempt, including reasoning and answers. The estimate excludes prompt prefill, context-dependent attention and serving overhead. It is an approximate compute comparison, not measured inference cost.^[1]

References

1. ^1 ^2 ^3 ^4 ^5 ^6 ^7 ^8 ^9 ^10 ^11Reflection. "Introducing Beam: Reflection's 501B open-weight model". October 5, 2026.
2. ^1 ^2 ^3 ^4 ^5Reflection Developer Docs. "Models". Accessed October 7, 2026.
3. ^1 ^2Reflection Developer Docs. "Reasoning". Accessed October 7, 2026.
4. ^Reflection Developer Docs. "Introduction". Accessed October 7, 2026.
5. ^Reflection Developer Docs. "Coding agent quickstart". Accessed October 7, 2026.
6. ^Reflection Developer Docs. "Tool calling". Accessed October 7, 2026.
7. ^Reflection Developer Docs. "Structured outputs". Accessed October 7, 2026.
8. ^DeepSeek-AI et al. "DeepSeek-V3 Technical Report," section 2.1.2. arXiv:2412.19437v2. February 18, 2025.