← All models
Reflection · Language model
Beam 1.0 501B A23B
reflection/beam-1.0-501b-a23b- Language
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
- Context window
- 1,048,576 tokens
- Max output
- 131,072 tokens
- Input
- text
- Output
- text
- Reasoning effort
- max, xhigh, high, medium, low (always on)
- Tokenizer
- Other
- Knowledge cutoff
- 2026-06-30
Supported parameters
Accepted by the model and forwarded as-is by the gateway.
- frequency_penalty
- 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
Use it
POST /api/v1/chat/completions · MCP chat · full reference
curl https://omnirail.org/api/v1/chat/completions \
-H "Authorization: Bearer $OMNIRAIL_KEY" -H "Content-Type: application/json" \
-d '{"model":"reflection/beam-1.0-501b-a23b","messages":[{"role":"user","content":"Hello"}],"max_tokens":300}'