Reflection: Beam 1.0 501B A23B
- by reflection
- reflection/beam-1.0-501b-a23b
Beam 1.0 is Reflection's open-weight reasoning model for coding and long-horizon agentic workflows. It uses a sparse mixture-of-experts architecture with 501B total and 23B active parameters per token, and a...
What the numbers mean
- Context window1,048,576 tokens
- How much the model can read in one request, counting your prompt, any files and the conversation so far. That is about 790,000 words, or 1,600 pages.
- Max output943,718 tokens
- The longest single reply it can write: about 710,000 words.
- Input price$0.3 per million tokens
- What you pay for everything the model reads. A 1,000-word prompt costs less than a cent.
- Output price$1 per million tokens
- What you pay for everything the model writes. Output usually costs more than input. A 500-word answer costs less than a cent.
- Modalitiestext in, text out
- The kinds of content it accepts and the kinds it produces. A model that takes images in can read a screenshot; one that only puts text out cannot draw.
- Knowledge cutoff2026-06-30
- The last date its training data covers. It does not know about anything after that unless you put it in the prompt.
- Intelligence index17.7
- One score that combines several reasoning, knowledge and coding benchmarks. Higher is better. This model ranks 68 of the 96 scored models listed here.
Supported parameters
The settings you can pass with a request. A model that lists tools can call functions; one that lists structured_outputs can be held to a JSON schema.