Beam 1.0 501B A23B

ReflectionMCP Tool CallReasoning
reflection/beam-1.0-501b-a23bOpen Weights

Sparse mixture-of-experts reasoning model with 23B active parameters for coding and agentic workflows

Benchmark Pricing$0.50 / 1M in$1.50 / 1M out$0.05 cache read
Context: 1,048,576 tokens•Max Output: 131,072 tokens•Knowledge: 2026-06-30•Released: 2026-10-05

TECHNICAL SPECIFICATIONS

Context Window1,048,5761M tokens
Max Output131,072131.1K tokens
LicenseApache-2.0Self-hostable
Input Modalities
Text
Function / Tool CallingSupported (MCP Ready)
Structured OutputsGuaranteed Schema

MODEL CONTEXT PROTOCOL (MCP) READINESS

✓ Beam 1.0 501B A23B natively supports OpenAI / Anthropic format tool calling and is fully compatible with MCP servers and agent loops.

MCP Context Window Recommendations

With a context capacity of 1,048,576 tokens, this model can handle extremely large MCP resource manifests, complete database schemas, and multiple active server tool definitions simultaneously without degradation.

STANDARDIZED BENCHMARKS & EVALS

DeepSWEUnit not reported
43.3
SWE-Bench ProUnit not reported
76.5
SWE-Bench ProUnit not reported
66.1
Terminal-BenchUnit not reported
80.1
SWE-Atlas Codebase QnAUnit not reported
31.8
SWE-Bench MultilingualUnit not reported
78.3
SWE-Bench VerifiedUnit not reported
81.6
AIME 2026Unit not reported
97.7
Humanity's Last ExamUnit not reported
36
Humanity's Last ExamUnit not reported
51.1
SciCodeUnit not reported
49.2
CritPtUnit not reported
16.3
GPQA DiamondUnit not reported
90.6
LiveCodeBenchUnit not reported
88.3
AutomationBenchUnit not reported
39.3
MCP AtlasUnit not reported
80.8
τ3 BankingUnit not reported
36.1
BrowseCompUnit not reported
76.8
DeepSearchQAUnit not reported
80.5
AA-LCRUnit not reported
78.2
LongBenchUnit not reported
67.5
IFBenchUnit not reported
79.6
AA-Omniscienceindex
10.8

API CALL INVOCATION EXAMPLE

// Example: Calling Beam 1.0 501B A23B via OpenAI SDK
import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.REFLECTION_API_KEY,
});

const response = await client.chat.completions.create({
  model: "reflection/beam-1.0-501b-a23b",
  messages: [{ role: "user", content: "Hello! Can you inspect MCP tools?" }],
});

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

PRICING BY PROVIDER

8 providers
ReflectionBeam-1.0-501B-A23B
$0.00 / $0.00$0.00 cache
NanoGPTreflection/beam-1.0
$0.50 / $1.50$0.05 cache
CoreWeavereflection/Beam-1.0-501B-A23B
$0.30 / $1.00$0.04 cache
Cloudflare AI Gatewayreflection-ai/beam-1.0-501b-a23b
$0.00 / $0.00$0.00 cache
Nebius Token Factoryreflection/beam
$0.25 / $1.00$0.25 cache
Together AIreflection/Beam-1.0-501B-A23B
$0.40 / $1.20$0.04 cache
Kilo Gatewayreflection/beam-1.0-501b-a23b
$0.25 / $1.00$0.05 cache
OpenRouterreflection/beam-1.0-501b-a23b
$0.30 / $1.00$0.04 cache

Rates in USD per 1M tokens. Rates synchronized from models.dev public catalog.

Prompt Caching Active

This model supports prompt caching at $0.05 / 1M tokens, saving up to 90% on repetitive MCP tool definitions and system instructions.

MODEL RESOURCES

View Reflection HubHugging Face Weights