Beam 1.0 501B A23B

reflection/beam-1.0-501b-a23b

reflection· 2026-10-05· 1.0M ctx

open weightstool callreasoningstructured quant: unknown

spread (output, p90 ÷ p10)

— too few providers

$1.20 – $3.00 · 2 providers · max÷min 2.5×

2 providers priced · 2 offerings

Including routers & resellers

Cheapest / median / spread use providers that serve the model on their own capacity: first-party lab, hyperscaler, inference host.

Provider classes in the stats:

Blended $/M assumptionsInput : output: 1Cache-hit share of input%

blended = 90.9% input × [(1 − 70% cache) × input + 70% cache × cache-read] + 9.1% × output + cache writes (30% of uncached input) where a write price is published

A provider that publishes no cache price is blended with cache share 0 (all input at the input price) and labelled “no cache price published” — never penalised against identical in/out prices.

Provider	Input	Output	Cache rd	Blended
Kilo Gateway router / aggregator	$0.25	$1.00	—	$0.32no cache price published
Together AI inference host	$0.40	$1.20	$0.04	$0.24
NanoGPT router / aggregator	$0.50	$1.50	$0.05	$0.30
OpenRouter router / aggregator	$1.00	$3.00	$0.10	$0.61
CoreWeave inference host	$1.00	$3.00	$0.10	$0.61

Unpriced · 2

Listed at $0 with no published pricing — we can't tell whether it's free, bundled or just missing. Excluded from cheapest, median and spread.

Cloudflare AI Gateway reflection-ai/beam-1.0-501b-a23bunpriced
Reflection Beam-1.0-501B-A23Bunpriced

Connect

Provider (default: cheapest output among direct providers)

$1.20 /M · inference host

opencode.json

{
  "provider": {
    "togetherai": {
      "npm": "@ai-sdk/togetherai",
      "models": {
        "reflection/Beam-1.0-501B-A23B": {}
      }
    }
  }
}

Set TOGETHER_API_KEY in your environment. Generated — verify against the OpenCode docs.

Not enough history yet — the min/median/max sparkline fills in as the daily snapshot accumulates.

Data: models.dev (MIT), snapshot 2026-10-17. Prices are vendor-published; verify before purchase. Methodology.