Beam 1.0
Beam 1.0 is capable in instruction following, reasoning, and coding. Too few results yet to rate agentic tasks, safety, long context, math, multimodal tasks, multilingual tasks, or factuality.
Capability profile
Bars show the model's median result as a share of the leading model's, per capability. Select a row to see its results.
Too few results yet to rate Agentic, Safety, Long Context, Math, Multimodal, Multilingual or Factuality.
Price
No current price is tracked for this model. See the rate card
Evidence
25results on19benchmarks
- 8 independently verified
- 17 vendor-reported
From 3 sources · latest Nov 4, 2026 · How verification works
Beam 1.0 benchmark results
25 results on 19 benchmarks, grouped by capability. Each bar is the result as a share of the capability leader's; every result links to its source.
20.0% behind the leader1 of 3 ranked benchmarks measured
- 79.60Nov 4, 2026
20.9% behind the leader4 of 6 ranked benchmarks measured
- GPQA Diamond90.60Nov 4, 2026GPQA
- 36.00Nov 4, 2026
- 16.30Nov 4, 2026
- 23.47Oct 20, 2026
22.0% behind the leader5 of 10 ranked benchmarks measured
- 80.10Nov 4, 2026
- 81.60Nov 4, 2026
- 78.30Nov 4, 2026
- 49.20Nov 4, 2026
- 66.10Nov 4, 2026
0 of 7 ranked benchmarks measured
- 76.80Nov 4, 2026
- 80.80Nov 4, 2026
0 of 3 ranked benchmarks measured
- 67.50Nov 4, 2026
0 of 5 ranked benchmarks measured
- 97.70Nov 4, 2026
More results
Benchmarks outside the capability baskets. They are not ranked against a leader.
- 45.00Oct 20, 2026
- 55.67Oct 20, 2026
- 67.67Oct 20, 2026
- 74.00Oct 20, 2026
- DeepSearchQA (F1)80.50Nov 4, 2026DeepSearchQA
- τ³-Bench Banking36.10Nov 4, 2026Tau3 Banking
Show 2 more resultsHide 2 results
- 43.30Nov 4, 2026
- HLE (with tools)51.10Nov 4, 2026Humanity's Last Exam (with tools)
Beam 1.0: common questions
Who makes Beam 1.0?
Beam 1.0 is made by Reflection.
What is Beam 1.0 good at?
Beam 1.0 is capable in instruction following, reasoning, and coding. Too few results yet to rate agentic tasks, safety, long context, math, multimodal tasks, multilingual tasks, or factuality.
How many benchmarks has Beam 1.0 been tested on?
We track 25 results for Beam 1.0 on 19 benchmarks from 3 sources, 8 of them independently verified. The latest was recorded on Nov 4, 2026.
About this record
Where Beam 1.0's numbers come from, and every name it appears under.
- Tracked since
- Oct 17, 2026
- Newest source mention
- Oct 20, 2026
Where the results come from
Verification: 25 scores · 8 independently verified · 17 vendor-reported. How these tiers are assigned
From 3 sources on 2 sites. api.llm-stats.com supplies 17 of them; the 8 independently verified results come from 1 site. Bars are coloured by trust tier.
- api.llm-stats.com17
- arcprize.org8
Also known as
How our sources name Beam 1.0 at each reasoning setting.
| Setting | Short form |
|---|---|
| low | beam 1.0 (low) |
| medium | beam 1.0 (medium) |
| high | beam 1.0 (high) |
| xhigh | beam 1.0 (xhigh) |