Beam 1.0 501B A23B
Beam 1.0 501B A23B is a large language model made by Reflection, added to the catalog on 16 Oct 2026. It has a context window of 1,048,576 tokens and costs $0.25 per 1M input tokens and $1.00 per 1M output tokens. It accepts text and produces text. Typical uses: reasoning over extensive text, coding with long text context and tool-assisted coding workflows.
Price checked ยท source: OpenRouter
What Beam 1.0 501B A23B is
Reflection: Beam 1.0 501B A23B is an open-weight reasoning model from Reflection, described as intended for coding and long-horizon agentic workflows. It accepts text and produces text. The catalog lists support for reasoning, tools, structured outputs, and response_format, making it suitable for text-based workflows involving tool interfaces or outputs constrained to a specified format. The catalog does not provide further details about tool behavior or supported schemas.
The model has a context window of 1,048,576 tokens and a maximum output of 131,072 tokens. This allows extensive text to be supplied as context and supports requests for long text outputs, within those limits. Its context capacity suits work involving lengthy source materials, extended coding discussions, or multi-stage tasks where relevant information needs to remain in one request. The output limit allows for substantial responses, depending on the task.
Its architecture is described as a sparse mixture of experts, with 501 billion total parameters and 23 billion active per token. The provided information does not specify supported programming languages, data sources, tool integrations, or evaluation results. Beam 1.0 is a text model to consider for reasoning and coding tasks that benefit from a large context and optional structured or tool-oriented interaction. The catalog lists no image, audio, or other non-text inputs or outputs.
Key features
- Accepts text and produces text
- Has a context window of 1,048,576 tokens
- Supports reasoning, tools, structured outputs, and response_format
- Uses a sparse mixture-of-experts architecture with 501B total and 23B active parameters
Typical uses
- Reasoning over extensive text
- Coding with long text context
- Tool-assisted coding workflows
- Generating structured text outputs
- Multi-stage agentic workflows
Beam 1.0 501B A23B pricing
US dollars, list price through OpenRouter.
| Item | Standard |
|---|---|
| Input, per 1M tokens | $0.25 |
| Output, per 1M tokens | $1.00 |
| Cached input read, per 1M tokens | $0.05 |
No price change since it was added on 16 Oct 2026.
| What a task costs | Cost |
|---|---|
| Answer a short question500 in / 300 out | $0.0004 |
| Summarise a 50-page PDF25,000 in / 800 out | $0.0071 |
| Review a 2,000-line source file30,000 in / 2,000 out | $0.0095 |
| Read a 300-page contract set150,000 in / 1,500 out | $0.039 |
Beam 1.0 501B A23B specifications
| Model id | reflection/beam-1.0-501b-a23b |
|---|---|
| Version | beam-1.0-501b-a23b-20261016 |
| Context window | 1,048,576 tokensAbout 786,432 English words |
| Max output | 131,072 tokens |
| Knowledge cutoff | 2026-06-30 |
| Added to catalog | 16 Oct 2026 |
| Input types | text |
|---|---|
| Output types | text |
| Tool calling | Yes |
| Structured outputs | Yes |
| Reasoning effort control | Yes |
| Content moderation | Not applied by provider |
Sample outputs
The same prompts are given to every model, and the unedited output is shown with the date it was produced. Upend does not publish leaderboard scores it did not run.
No output has been recorded for Beam 1.0 501B A23B yet.