Leaderboard · #30 of 33 · Reflection · Run it yourself

Beam 1.0

A new lab's open model, free to use commercially, but 18 on Artificial Analysis's index is within noise of Qwen3.8-27B, a model small enough for one workstation.

Updated Nov 12, 2026 · kept by WrenJSON

Key facts

Score
43 / 100, rank #30
Tests28Pricen/aPeoplen/aFit85
Price
No verified list price
Reads at once
1M tokens (about 750,000 words)
Where you can get it
Amazon SageMaker JumpStart
Host it yourself
Yes (open weights)
Released
Oct 16, 2026
Best for
Running AI on your own servers, Companies that need a standard open licence

Reflection has not published a list price for Beam 1.0 that we could verify from a primary source. See what every model costs for the same job.

Independent tests

TestResultWho ran itRead
Artificial Analysis Intelligence Index18Artificial Analysis LLM leaderboardNov 12, 2026

Caveats

  • There is no per-use price, and it is a very large model that needs data-centre hardware.
  • Text only, and built mainly for English, per its model card.
  • One independent test; Reflection's own results are not counted.
  • On Amazon SageMaker JumpStart it launched in two US regions only (N. Virginia and Ohio), per AWS.

Questions people ask

How much does Beam 1.0 cost?

Reflection has not published a list price for Beam 1.0 that we could verify from a primary source.

How good is Beam 1.0?

Beam 1.0 ranks #30 of 33 on the Models at Work leaderboard with a score of 43 out of 100 (November 2026 edition). A new lab's open model, free to use commercially, but 18 on Artificial Analysis's index is within noise of Qwen3.8-27B, a model small enough for one workstation.

What is Beam 1.0 best for?

Beam 1.0 is best for running ai on your own servers, companies that need a standard open licence.

How does Beam 1.0 do on independent tests?

Artificial Analysis Intelligence Index: 18 (Artificial Analysis LLM leaderboard).

How much can Beam 1.0 read at once?

Beam 1.0 can take in 1M tokens (about 750,000 words) in one go, according to Reflection. Vendors call this the context window.

Where can a business get Beam 1.0?

Beam 1.0 is available through Amazon SageMaker JumpStart.

Can a company host Beam 1.0 itself?

Yes. Beam 1.0 is released with open weights, which means a company can download the model and run it on its own servers. That needs engineers.

Compare Beam 1.0

    How the score works: Score is 0 to 100 and computed, not typed. 40% is independent testing (Artificial Analysis, Arena, Scale, Epoch, Vals and others; each test is scaled against the best score on this board, and tests of real work, such as finished freelance projects or professional tasks, count double the exams). 30% is price: the vendor's list price for a typical mix of reading and writing, on a fixed scale from $0.05 to $60 per million tokens, so a model that costs a fifth as much gains real ground. 15% is what people report after running it, from Signals and linked threads. 15% is fit: how much text it can take in at once and where a business can get it (the products and clouds that carry it), with a small credit for open weights. A missing component drops out and the rest are reweighted. A model with no independent test yet is provisional and ranks below every measured one. Within three points is a tie. Vendor-published figures are labelled and never counted. Our coverage: read the article.