impulse

Beam 1.0 501B A23B

Open weights

Beam 1.0 is Reflection's open-weight reasoning model for coding and long-horizon agentic workflows.

API id Beam-1.0-501B-A23BReleased 16 Oct 2026 (5 weeks ago)
Input / 1M
$0.30
at W&B Inference
Output / 1M
$1.00
at W&B Inference
Context window
1M
tokens
Max output
1M
tokens per reply
Hosts
2
2 listings
Released
5 weeks ago
16 Oct 2026

Pricing

List prices at W&B Inference, in US dollars per million tokens.

TierInput / 1MOutput / 1M
Standard$0.30$1.00
Cached input
87% off input
$0.04

What it costs a month

At W&B Inference list prices, 30 days.

  • Chat assistant5,000 requests a day, 2K in / 0.5K out, cached$142
  • RAG over documents2,000 requests a day, 8K in / 0.6K out, cached$118
  • Coding agent1,000 requests a day, 40K in / 2K out, cached$202
  • Overnight batch20,000 requests a day, 4K in / 0.4K out$960
Use your own numbers

Where to run it

2 hosts serve Beam 1.0 501B A23B. The comparison is on blended price, 3 parts input to 1 part output.

HostIn / out per 1M
W&B InferenceInference
Beam-1.0-501B-A23B
$0.30 / $1.00
Together AIInference
Beam-1.0-501B-A23B
$0.40 / $1.20
+26% vs headline

Capabilities

Core

  • Streaming
  • Function calling
  • JSON mode
  • Structured outputs

Multimodal

  • Vision
  • Image input
  • Audio
  • Image generation
  • Video

Advanced

  • Reasoning
  • Tool calling
  • MCP
  • Prompt caching
  • Batch API
  • Embeddings
Takes in
Text
Produces
Text
Features
Prompt caching
Open weights
License: Apache-2.0
Runs in about 1203 GB of GPU memory

Cheaper alternatives

Current models that cost less and keep most of what this one can do: tools, vision, reasoning, structured output, caching and enough context.

Same model, cheaper host

No host sells it for less than W&B Inference.

Smaller Independent models

No cheaper Independent model keeps its key capabilities.

Other frontier labs

  • GPT-5.4 Nano
    OpenAI · on OpenAI
    $0.20 / $1.25
    Has every key capability
    3% less
  • Claude Haiku 5.5
    Anthropic · on Anthropic
    $0.10 / $0.50
    Has every key capability
    58% less
  • GPT-6 Luna
    OpenAI · on OpenAI
    $0.10 / $0.50
    Has every key capability
    58% less
  • GPT-4o Mini
    OpenAI · on OpenAI
    $0.15 / $0.60
    Lacks reasoning
    45% less

Open-weight picks

  • MiMo V2.6 Pro
    Xiaomi · on GMI Cloud
    $0.348 / $0.696
    Has every key capability
    8% less
  • Muse Glimmer 30B
    Meta · on EmpirioLabs AI
    $0.20 / $0.80
    Has every key capability
    26% less
  • Deepseekv4.1 Flash
    DeepSeek · on Sarvam
    $0.149 / $0.585
    Has every key capability
    46% less
  • Qwen3.8 27B
    Alibaba Qwen · on EmpirioLabs AI
    $0.17 / $0.50
    Has every key capability
    47% less

Call it

Using W&B Inference's API.

Python
from litellm import completion

response = completion(
    model="wandb/Beam-1.0-501B-A23B",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

Quick answers

How much does Beam 1.0 501B A23B cost?

On W&B Inference, Beam 1.0 501B A23B costs $0.30 per million input tokens and $1.00 per million output tokens. Cached input costs $0.04 per million tokens.

Which APIs offer Beam 1.0 501B A23B?

Beam 1.0 501B A23B runs on 2 APIs: W&B Inference and Together AI.

What is Beam 1.0 501B A23B's context window?

1M tokens, with up to 1M tokens in one reply.

Is Beam 1.0 501B A23B being retired?

No retirement date is scheduled for it on any API we track.

What are cheaper alternatives to Beam 1.0 501B A23B?

Models that cost less and keep most of what it can do: GPT-5.4 Nano on OpenAI (3% less), Claude Haiku 5.5 on Anthropic (58% less) and MiMo V2.6 Pro on GMI Cloud (8% less). Savings are on a mix of 3 input tokens to 1 output token.

When was Beam 1.0 501B A23B released?

Independent released it on 16 Oct 2026.

What is Beam 1.0 501B A23B's API model id?

Use Beam-1.0-501B-A23B with W&B Inference.

Sources

Raw catalog entry for Beam-1.0-501B-A23B on W&B Inference
{
  "cache_read_input_token_cost": 4e-8,
  "input_cost_per_token": 3e-7,
  "litellm_provider": "wandb",
  "max_input_tokens": 1049000,
  "mode": "chat",
  "output_cost_per_token": 0.000001,
  "source": "https://www.coreweave.com/products/coreweave-forge/serverless-inference",
  "supports_function_calling": true,
  "supports_prompt_caching": true,
  "supports_reasoning": true,
  "supports_response_schema": true,
  "supports_tool_choice": true,
  "supports_vision": false
}