Reflection · Official-source reference

Reflection Beam 1.0 501B A23B

Hosted API reference
$0.25 / $1
Nebius via OpenRouter · input / output per million tokens · Nebius hosted offer, not a Reflection first-party tariff. Checked October 17, 2026.
Context
1M
Output · USD · Output · Hosted API prices · USD per 1M tokens
Nebius via OpenRouter$1
$1
Together via OpenRouter$1.2
$1.2

Reflection Beam 1.0 501B A23B · Published pricing

Hosted API prices · USD per 1M tokens
ProviderInputOutputCached inputScope
ReflectionNot announcedNot announcedNot announcedFirst-party public tariff not found
Nebius via OpenRouter$0.25$1.00Not announcedHosted-provider offer
Together via OpenRouter$0.40$1.20$0.04Hosted-provider offer

Reflection Beam 1.0 501B A23B · Weight license

Weight license
Model weightsLicense
Open weightsApache 2.0

Reflection Beam 1.0 501B A23B · Weights and quantization

Apache 2.0 deployment and fine-tuning
ItemConfiguration / evidence
WeightsBF16 / FP8 / NVFP4; official Hugging Face checkpoints

Reflection Beam 1.0 501B A23B · Deployment options

Apache 2.0 deployment and fine-tuning
ItemConfiguration / evidence
Self-hostingDownload weights → SGLang / vLLM / TokenSpeed → Beam Verifier

Reflection Beam 1.0 501B A23B · Fine-tuning support

Apache 2.0 deployment and fine-tuning
ItemConfiguration / evidence
Fine-tuningLoRA / QLoRA / SFT / RL; stack-specific recipes coming soon

Reflection Beam 1.0 501B A23B · Cost examples

Same workload · 1M input + 200K output
ProviderInput costOutput costTotal USD
Nebius$0.25$0.20$0.45
Together$0.40$0.24$0.64

Reflection Beam 1.0 501B A23B · Known risks and caveats

Apache 2.0 deployment and fine-tuning
ItemConfiguration / evidence
EfficiencyReflection reports about 3× fewer reasoning tokens; no matched GLM/Kimi efficiency result verified
Evidence limitsVendor benchmarks use their own harnesses; long-horizon production stability is not established

Reflection Beam 1.0 501B A23B · Use cases

Apache 2.0 deployment and fine-tuning
ItemConfiguration / evidence
Agent integrationMirror CLI; coding, terminal tasks and tool use

Cost examples

  • Beam is Reflection’s Apache 2.0 text model with 501.3B total and 23.4B active parameters.

Context and limits

  • Reflection Beam-1.0-501B-A23B has no first-party token tariff in the reviewed public documentation.

Questions about Reflection Beam 1.0 501B A23B pricing