reflectionNot yet certifiedReleased 10/16/2026

Beam 1.0

Beam is Reflection’s 501-billion-parameter mixture-of-experts model with 23 billion active parameters, built for coding, reasoning, and agentic workloads. It supports text generation, tool calling, structured outputs, and five reasoning effort levels, with a 262,144-token context window and up to 131,072 output tokens.

View API reference ↓
Input / output priceFrom$0 / $0per 1M
24h uptime99.62%

Listed for reference. Inference Bureau dispatches only to certified routes, so requests for this model are refused until one is certified. See certified models.

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.inferencebureau.com/v1",
  apiKey: process.env.IB_API_KEY,
});

const res = await client.chat.completions.create({
  model: "reflection/beam",
  messages: [{ role: "user", content: "Why is the sky blue?" }],
});

console.log(res.choices[0].message.content);

Providers

Dispatch routes requests across these providers and fails over between them without changing the model. Restrict or order providers with team Directives. Prices are gateway list prices before the platform fee.

ProviderContextMax outputLatencyThroughputInputOutputCacheCapabilitiesZDRNo trainingRegionsStatus
Baseten1.05M131K0.1 s—$0.40$1.20Read $0.040Write —5 featuresYesYesGlobalListed
nebius1.05M1.05M0.3 s557 tps$0.25$1.00Read —Write —3 featuresYesYesGlobalListed
reflection1.05M131K0.5 s215 tps$0$0Read $0Write —6 featuresNoNoGlobalListed
Together AI1.05M1.05M0.1 s—$0.40$1.20Read $0.040Write —5 featuresYesYesGlobalListed

Uptime

24 hours

Request success rate per provider.

Baseten99.80%
nebius99.94%
reflection98.81%
Together AI99.92%

Throughput

last hour

P50 output tokens per second.

nebius557 tps
reflection215 tps
Baseten—
Together AI—

Latency

last hour

P50 time to first token.

Together AI0.1 s
Baseten0.1 s
nebius0.3 s
reflection0.5 s

Getting started

Create a key on the API Keys page and set it as IB_API_KEY. See the quickstart.

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.inferencebureau.com/v1",
  apiKey: process.env.IB_API_KEY,
});

const res = await client.chat.completions.create({
  model: "reflection/beam",
  messages: [{ role: "user", content: "Why is the sky blue?" }],
});

console.log(res.choices[0].message.content);

Top-level parameters

Standard Chat Completions parameters work as documented. These have model-specific behavior.

ParameterTypeRequiredDescription
modelstringYesModel ID in the form creator/model, e.g. reflection/beam. Must be allowed by your team's Directives.
max_tokensnumberNoHard cap on generated tokens, and the basis for the balance reserved per request. Beam 1.0 supports up to 131,072 output tokens.
reasoning{ effort: 'low' | 'medium' | 'high' }NoReasoning effort, mapped to the provider's native configuration. Reasoning tokens count toward output usage.
toolsTool[]NoFunction definitions the model can call. Tool calls are returned in choices[].message.tool_calls.
Supportedmax_tokenstemperaturestoptoolstool_choicereasoninginclude_reasoningresponse_formatstructured_outputs

Input limits

InputSourcesLimits
Text—Prompt and response share the 1.05M-token context window

More models by reflection

ModelContextInputOutputCacheRelease
reflection/beam-free1.05M$0$0Read —Write —10/16/2026