Beam 1.0 501B A23B

Model ID

Use with the Workers AI binding, the AI Gateway REST API, or the Unified API.

Calling the provider directly? Use the provider's model name instead, including with provider-native endpoints or the provider's own SDK. For example, Anthropic's name for anthropic/claude-opus-4.8 is claude-opus-4-8.

Third-party
Zero data retention

Beam-1.0-501B-A23B is Reflection AI's open-weights mixture-of-experts model with 501B total parameters and 23B active parameters, built for coding, reasoning, and agentic workloads.

Model Info
Context Window ↗	262,144 tokens
Terms and License	link ↗
More information	link ↗
Zero data retention	Yes
Request formats	Chat Completions
Pricing	Input (per 1M tokens)$0.00 Cached input (per 1M tokens)$0.00 Output (per 1M tokens)$0.00

Usage

const response = await env.AI.run(
  'reflection-ai/beam-1.0-501b-a23b',
  {
    messages: [{ role: 'user', content: 'Explain why the sky is blue in one sentence.' }],
    max_completion_tokens: 256,
    reasoning_effort: 'low',
  },
)
console.log(response)

curl https://api.cloudflare.com/client/v4/accounts/$CLOUDFLARE_ACCOUNT_ID/ai/v1/chat/completions \
  --header "Authorization: Bearer $CLOUDFLARE_API_TOKEN" \
  --header "Content-Type: application/json" \
  --data '{
  "model": "reflection-ai/beam-1.0-501b-a23b",
  "messages": [
    {
      "role": "user",
      "content": "Explain why the sky is blue in one sentence."
    }
  ],
  "max_completion_tokens": 256,
  "reasoning_effort": "low"
}'

The sky is blue because sunlight undergoes Rayleigh scattering in the atmosphere, which preferentially redirects shorter blue wavelengths toward your eyes.

{
  "id": "chatcmpl-002d24bd-de9c-4bab-947d-47f77166eb5c",
  "object": "chat.completion",
  "created": 1792089654,
  "model": "reflection-ai/beam-1.0-501b-a23b",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "The sky is blue because sunlight undergoes Rayleigh scattering in the atmosphere, which preferentially redirects shorter blue wavelengths toward your eyes.",
        "refusal": null,
        "reasoning_content": "The user asks: \"Explain why the sky is blue in one sentence.\" Straightforward. Need one sentence. Ensure accuracy: Rayleigh scattering scatters shorter blue wavelengths more. Provide concise. No safety issues."
      },
      "finish_reason": "stop",
      "logprobs": null
    }
  ],
  "usage": {
    "prompt_tokens": 506,
    "completion_tokens": 79,
    "total_tokens": 585,
    "prompt_tokens_details": {
      "cached_tokens": 448
    },
    "completion_tokens_details": {
      "reasoning_tokens": 50
    }
  }
}

Examples

Structured project status — Return a typed project status summary using structured output

const response = await env.AI.run(
  'reflection-ai/beam-1.0-501b-a23b',
  {
    messages: [
      {
        role: 'user',
        content:
          'Summarize this project update: API integration is complete, security review is scheduled for Friday, and the mobile client is blocked on final design assets.',
      },
    ],
    max_completion_tokens: 512,
    response_format: {
      type: 'json_schema',
      json_schema: {
        name: 'project_status',
        strict: true,
        schema: {
          type: 'object',
          properties: {
            overall_status: { type: 'string', enum: ['on_track', 'at_risk', 'blocked'] },
            completed: { type: 'array', items: { type: 'string' } },
            next_steps: { type: 'array', items: { type: 'string' } },
            blockers: { type: 'array', items: { type: 'string' } },
          },
          required: ['overall_status', 'completed', 'next_steps', 'blockers'],
          additionalProperties: false,
        },
      },
    },
  },
)
console.log(response)

curl https://api.cloudflare.com/client/v4/accounts/$CLOUDFLARE_ACCOUNT_ID/ai/v1/chat/completions \
  --header "Authorization: Bearer $CLOUDFLARE_API_TOKEN" \
  --header "Content-Type: application/json" \
  --data '{
  "model": "reflection-ai/beam-1.0-501b-a23b",
  "messages": [
    {
      "role": "user",
      "content": "Summarize this project update: API integration is complete, security review is scheduled for Friday, and the mobile client is blocked on final design assets."
    }
  ],
  "max_completion_tokens": 512,
  "response_format": {
    "type": "json_schema",
    "json_schema": {
      "name": "project_status",
      "strict": true,
      "schema": {
        "type": "object",
        "properties": {
          "overall_status": {
            "type": "string",
            "enum": [
              "on_track",
              "at_risk",
              "blocked"
            ]
          },
          "completed": {
            "type": "array",
            "items": {
              "type": "string"
            }
          },
          "next_steps": {
            "type": "array",
            "items": {
              "type": "string"
            }
          },
          "blockers": {
            "type": "array",
            "items": {
              "type": "string"
            }
          }
        },
        "required": [
          "overall_status",
          "completed",
          "next_steps",
          "blockers"
        ],
        "additionalProperties": false
      }
    }
  }
}'

{"overall_status": "on_track",
"completed": ["API integration"],
"next_steps": ["Security review scheduled for Friday"],
"blockers": ["Mobile client awaiting final design assets"]}

{
  "id": "chatcmpl-781353f2-706b-4d56-b771-c785968bbf97",
  "object": "chat.completion",
  "created": 1792089661,
  "model": "reflection-ai/beam-1.0-501b-a23b",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "{\"overall_status\": \"on_track\",\n\"completed\": [\"API integration\"],\n\"next_steps\": [\"Security review scheduled for Friday\"],\n\"blockers\": [\"Mobile client awaiting final design assets\"]}",
        "refusal": null,
        "reasoning_content": "The user wants a summary of a project update. Straightforward. Just summarize."
      },
      "finish_reason": "stop",
      "logprobs": null
    }
  ],
  "usage": {
    "prompt_tokens": 527,
    "completion_tokens": 64,
    "total_tokens": 591,
    "prompt_tokens_details": {
      "cached_tokens": 512
    },
    "completion_tokens_details": {
      "reasoning_tokens": 18
    }
  }
}

Weather tool call — Choose a weather lookup function and supply its arguments

const response = await env.AI.run(
  'reflection-ai/beam-1.0-501b-a23b',
  {
    messages: [{ role: 'user', content: 'What is the weather in Lisbon today?' }],
    max_completion_tokens: 256,
    tool_choice: 'required',
    tools: [
      {
        type: 'function',
        function: {
          name: 'get_weather',
          description: 'Look up the current weather for a city.',
          parameters: {
            type: 'object',
            properties: {
              city: { type: 'string' },
              units: { type: 'string', enum: ['celsius', 'fahrenheit'] },
            },
            required: ['city', 'units'],
            additionalProperties: false,
          },
        },
      },
    ],
  },
)
console.log(response)

curl https://api.cloudflare.com/client/v4/accounts/$CLOUDFLARE_ACCOUNT_ID/ai/v1/chat/completions \
  --header "Authorization: Bearer $CLOUDFLARE_API_TOKEN" \
  --header "Content-Type: application/json" \
  --data '{
  "model": "reflection-ai/beam-1.0-501b-a23b",
  "messages": [
    {
      "role": "user",
      "content": "What is the weather in Lisbon today?"
    }
  ],
  "max_completion_tokens": 256,
  "tool_choice": "required",
  "tools": [
    {
      "type": "function",
      "function": {
        "name": "get_weather",
        "description": "Look up the current weather for a city.",
        "parameters": {
          "type": "object",
          "properties": {
            "city": {
              "type": "string"
            },
            "units": {
              "type": "string",
              "enum": [
                "celsius",
                "fahrenheit"
              ]
            }
          },
          "required": [
            "city",
            "units"
          ],
          "additionalProperties": false
        }
      }
    }
  ]
}'

{
  "id": "chatcmpl-fda9d6ad-dcb0-4b60-9e5a-4d78f406ecc1",
  "object": "chat.completion",
  "created": 1792089672,
  "model": "reflection-ai/beam-1.0-501b-a23b",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": null,
        "tool_calls": [
          {
            "id": "call-c105de7f-4e34-4ce9-bab9-ef2e6a89b645",
            "type": "function",
            "function": {
              "name": "get_weather",
              "arguments": "{\"city\": \"Lisbon\", \"units\": \"celsius\"}"
            }
          }
        ],
        "refusal": null,
        "reasoning_content": "The user is asking about the weather in Lisbon today. I have a tool to get weather. Let me call it with city Lisbon and units... the user didn't specify units. Celsius is reasonable default, but maybe I should pick. Let me use celsius as default. Actually, let me just call."
      },
      "finish_reason": "tool_calls",
      "logprobs": null
    }
  ],
  "usage": {
    "prompt_tokens": 544,
    "completion_tokens": 92,
    "total_tokens": 636,
    "prompt_tokens_details": {
      "cached_tokens": 512
    },
    "completion_tokens_details": {
      "reasoning_tokens": 69
    }
  }
}

Scheduling reasoning — Reason through a scheduling constraint with system guidance

const response = await env.AI.run(
  'reflection-ai/beam-1.0-501b-a23b',
  {
    messages: [
      {
        role: 'system',
        content: 'Answer concisely and show the scheduling logic in two bullet points.',
      },
      {
        role: 'user',
        content:
          'Maya is available Monday and Wednesday. Theo is available Wednesday and Thursday. Priya is available Monday, Wednesday, and Friday. Which day should they schedule a 90-minute meeting, and why?',
      },
      { role: 'user', content: 'The meeting must include all three attendees.' },
      { role: 'user', content: 'Please give the recommendation.' },
    ],
    max_completion_tokens: 384,
    reasoning_effort: 'medium',
  },
)
console.log(response)

curl https://api.cloudflare.com/client/v4/accounts/$CLOUDFLARE_ACCOUNT_ID/ai/v1/chat/completions \
  --header "Authorization: Bearer $CLOUDFLARE_API_TOKEN" \
  --header "Content-Type: application/json" \
  --data '{
  "model": "reflection-ai/beam-1.0-501b-a23b",
  "messages": [
    {
      "role": "system",
      "content": "Answer concisely and show the scheduling logic in two bullet points."
    },
    {
      "role": "user",
      "content": "Maya is available Monday and Wednesday. Theo is available Wednesday and Thursday. Priya is available Monday, Wednesday, and Friday. Which day should they schedule a 90-minute meeting, and why?"
    },
    {
      "role": "user",
      "content": "The meeting must include all three attendees."
    },
    {
      "role": "user",
      "content": "Please give the recommendation."
    }
  ],
  "max_completion_tokens": 384,
  "reasoning_effort": "medium"
}'

- All three must attend, so the meeting day must be in the intersection of Maya's (Mon, Wed), Theo's (Wed, Thu), and Priya's (Mon, Wed, Fri) calendars; only Wednesday appears for all three.
- Therefore, schedule the 90-minute meeting on Wednesday, since it is the only day all attendees are available.

{
  "id": "chatcmpl-26cd51e7-33f7-4048-93f0-69f51b1c66cc",
  "object": "chat.completion",
  "created": 1792089681,
  "model": "reflection-ai/beam-1.0-501b-a23b",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "- All three must attend, so the meeting day must be in the intersection of Maya's (Mon, Wed), Theo's (Wed, Thu), and Priya's (Mon, Wed, Fri) calendars; only Wednesday appears for all three.\n- Therefore, schedule the 90-minute meeting on Wednesday, since it is the only day all attendees are available.",
        "refusal": null,
        "reasoning_content": "The user asks: Maya available Mon/Wed, Theo available Wed/Thu, Priya available Mon/Wed/Fri. Meeting must include all three. Need schedule 90-min meeting. Which day should schedule and why? Concise, two bullet points showing scheduling logic. Constraints: all three must attend. Intersection: Maya ∩ Theo = Wed; Priya ∩ {Wed} = Wed. So Wednesday is common day. Need 90-minute meeting—presumably available all day? No time specifics. So recommend Wednesday; only day all three available. Answer with two bullet points.\n\nNeed ensure not violate policy. Fine.\n\nLet's craft concise."
      },
      "finish_reason": "stop",
      "logprobs": null
    }
  ],
  "usage": {
    "prompt_tokens": 572,
    "completion_tokens": 218,
    "total_tokens": 790,
    "prompt_tokens_details": {
      "cached_tokens": 448
    },
    "completion_tokens_details": {
      "reasoning_tokens": 138
    }
  }
}

Parameters

▶messages[]

arrayrequired

temperature

numberminimum: 0maximum: 2

max_tokens

numberexclusiveMinimum: 0

max_completion_tokens

numberexclusiveMinimum: 0

top_p

numberminimum: 0maximum: 1

frequency_penalty

numberminimum: -2maximum: 2

presence_penalty

numberminimum: -2maximum: 2

stream

boolean

▶stream_options{}

object

▶tools[]

array

tool_choice

response_format

▶modalities[]

array

▶audio{}

object

reasoning_effort

stringOptional reasoning control; availability and accepted values are model-dependent.

string

object

string

created

number

model

string

▶choices[]

array

▶usage{}

object

API Schemas (Raw)

Input

Output