The Orchestrator › Models › 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...
Specifications
| Provider | Reflection |
| Model ID | reflection/beam-1.0-501b-a23b |
| Context window | 1.0M tokens (1,048,576) |
| Modalities | text → text |
| Tokenizer | Other |
| Router sources | OpenRouter, TrustedRouter |
| Released | 2026-10-16 |
| Tracked since | 2026-11-14 |
| Popularity | Not in the ranked set |
| Reasoning | Mandatory |
Pricing (per million tokens)
| Variant | Source | API ID | Input | Output | Context |
|---|---|---|---|---|---|
| standard | OpenRouter | reflection/beam-1.0-501b-a23b | $0.3 | $1.00 | 1.0M |
| standard | TrustedRouter | reflection/beam-1.0-501b-a23b | $0.369 | $1.18 | 1.0M |
Recorded price history
| Date | Input / 1M | Output / 1M | Context |
|---|---|---|---|
| 2026-11-14 | $0.369 | $1.18 | 1.0M |
| 2026-11-14 | $0.3 | $1.00 | 1.0M |
Capabilities
Agentic tool useStructured outputsReasoning (always on)1M+ context windowControllable reasoningNo content storage (attested)
Trust flags come from TrustedRouter's per-provider attestations.
Supported parameters
frequency_penalty, include_reasoning, logit_bias, logprobs, max_tokens, min_p, presence_penalty, reasoning, reasoning_effort, repetition_penalty, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_k, top_logprobs, top_p