Skip to content

Reflection · Chat

Beam 1.0 501B A23B

ReflectionChatBeam familyOpen weights. One of its sources says the weights are publicly available.
From Aiml API's listing:
Beam 1.0 is Reflection's open-weight reasoning model for coding and long-horizon agentic workflows. A sparse mixture-of-experts with 501B total and 23B active parameters per token, it always reasons before answering, supports tool calling and structured outputs, and works over a 1M-token context.
llminfo IQ
130.0
Market median price
$0.60 blended / 1M
Max context
1M
Listings
27

Evidence and sources

🤗 reflection/Beam-1.0-501B-A23Bllminfo IQ 130.0, ranked among chat models. From 84 benchmarks. Plenty of data behind it.How llminfo IQ is computedCoding IQ 109.0, 95% interval 103–115. From 10 benchmark families across 2 sources. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.How Coding IQ is computed Training-data cutoff, as its sources state it
All rankings: #84 on OpenRouter today #93 on OpenRouter this week #111 on OpenRouter this monthSources: LMArena data, Models.dev data, AI/ML API data, OpenRouter data docsPrimary data from LMArena from Models.dev: weights availability from AI/ML API: primary mode, description from OpenRouter: knowledge cutoff

Advantages and trade-offs

Compared with 10 chat models, picked mainly by popularity and IQ

Advantages

Open weights available

The weights are public. 3 of the 9 peers that say either way have open weights.

We only make a claim when there's enough data on models of the same type. Updated 22 Nov, 06:07 UTC.Prices come from the model's maker where they sell it, otherwise their official partner, otherwise the market median. This comparison uses 8 maker's prices; 2 market median prices.

Input & output

as stated by the model's providers
Text
Text

Capabilities

Supported operations and provider-stated features

ReasoningFunction callingStructured outputPrompt caching

Listings

 Hugging FaceInference by Nebius AI Studioreflection/Beam-1.0-501B-A23BSource: Hugging Face Inference Providers data1M$0.25$1.00——
 Kiloreflection/beam-1.0-501b-a23bSources: Kilo AI Gateway data, Models.dev data1M$0.25cache read $0.05$1.00——
 Nous Portalreflection/beam-1.0-501b-a23bSource: Nous Portal Inference Gateway data1M$0.25cache read $0.05$1.00——
 OpenRouterInference by Nebius AI Studioreflection/beam-1.0-501b-a23b · fp4Source: OpenRouter data docs1M$0.25cache read $0.05$1.0083917ms
 Weights & Biasesreflection/Beam-1.0-501B-A23BSources: LiteLLM data, Models.dev data, Weights & Biases Inference data1M$0.30cache read $0.04$1.00——
 Ionetreflection/Beam-1.0-501B-A23BSource: IO Intelligence (io.net) data1M$0.35cache read $0.175$1.120.001424.333ms
 Eden AIInference by Together AItogether_ai/reflection/Beam-1.0-501B-A23BSource: Models.dev data1M$0.40cache read $0.04$1.20——
 Fastrouterreflection/beam-1.0-501b-a23bSource: FastRouter data1M$0.40cache read $0.04$1.20——
 Llmgatewaybeam-1.0-501b-a23bSource: LLMGateway data1M$0.40cache read $0.04$1.20——
 Wavespeedreflection/beam-1.0-501b-a23bSource: WaveSpeed AI data1M$0.40cache read $0.04$1.20——
Show all 23 listings

Showing 10 of 23: the cheapest per 1M input tokens from each provider.

Also sold by Infersia, Llmgateway Providers, Together AI.

4 listings without a price are hidden.

Benchmarks

Named effort levels run from lowest to highest. Select a result to see its source details.

Human preference

Agentic

Arena AgentSource: LMArena data · Nov 2026
MCP AtlasSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim

Coding

Arena ELOSource: LMArena data · Nov 2026
SciCodeSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim

Reasoning

Instruction following

IFBenchSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim

Long context

Math

Science

CritPTSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
GPQASource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim

Other

Data credits: Data by LLM Stats · Hugging Face official benchmark leaderboards; licence varies by benchmark · LMArena, CC BY 4.0 (licence)

Also known as: reflection/beam-1.0-501b-a23b, together-ai/beam-1.0-501b-a23b, together_ai/reflection/beam-1.0-501b-a23b, unknown/beam-1.0-501b-a23b, wandb/reflection/beam-1.0-501b-a23b

Like this one. Tap or hover any dotted value, label or icon to see where it comes from and how it is measured. open details when tapped or hovered. How this data works