BenchLeader

Beam 1.0 501B A23B

Reasoning effort
Quality#112 of 339
54.2Index
Above the median
+4.3 on the median
Price#101 of 255
$0.600per 1M tokens
Cheaper than most
0.7× the median
Speed#14 of 215
202tokens/s
Top 10%
2.6× the median
Value#94 of 255
90.3points per $
Above the median
1.7× the median

Better value: MiMo-V2.5 scores 56.3 for $0.138/M, a quarter of the price.

20 cheaper ranked models score higher: DeepSeek V4.1 Flash (58.5, $0.525), GPT-5.6 Luna (58.2, $0.450), DeepSeek V4.1 Flash 1017 (58.0, $0.525) and 17 more.

As of 15 Nov 2026, Beam 1.0 501B A23B by Reflection scores 54.2 on the BenchLeader Index (611 ranked configurations), with its best results in maths (65) and weakest in agents & tools (41). API pricing: $0.400 per 1M input tokens and $1.20 per 1M output tokens ($0.600 blended at 3:1), list price on 15 Nov 2026. It streams about 200 tokens/s (OpenRouter). Released 5 Oct 2026.

Key facts (as of 15 Nov 2026)

  • Beam 1.0 501B A23B API pricing: $0.400 per 1M input tokens and $1.20 per 1M output tokens ($0.600 blended at 3:1) (list price, 15 Nov 2026).
  • Beam 1.0 501B A23B streams about 200 tokens/s (OpenRouter, 15 Nov 2026).
  • Beam 1.0 501B A23B was released on 5 Oct 2026 (Reflection).

In detail. Beam 1.0 501B A23B is a Reflection open-weights reasoning model, released 5 Oct 2026. Its best configuration ranks #236 of 611 on the BenchLeader Index at 54.2 ±7.6, in the upper half. The ± is the point: 380 other configurations score within that range, so they and this one cannot be told apart on quality alone — price and speed are what separate them. It scores highest in maths (65) and lowest in agents & tools (41). At $0.600 per million tokens blended it is mid-priced. Output speed of 202 tokens per second puts it in the fastest quarter, with a first token in 0.3 s. It has been measured at 4 reasoning-effort settings; this summary describes the best-scoring one, and the tabs above switch between them. Last measured 11 Nov 2026.

Blended price
$0.600/M
$0.400 in · $1.20 out
via models.dev
Output speed
202 tok/s
OpenRouter traffic, 7-day median; not yet measured by Artificial Analysis
First answer
0.34 s
Context
1.0M
via OpenRouter
Cost per task
–
Tokens per task
–
Released
5 Oct 2026
How it scores by categoryDashed line = average model (50). One step of 15 = one standard deviation.
  1. Overall index54
  2. Reasoning62
  3. Coding52
  4. Agents & tools41
  5. Maths65
  6. Human preference61

Embed this rank

More badge styles
[![Beam 1.0 501B A23B on BenchLeader: #113 overall](https://www.benchleader.com/badge/beam-1-0-501b-a23b.svg?style=compact)](https://www.benchleader.com/models/beam-1-0-501b-a23b)

Paste it at the top of README.md, beside your other badges. The badge redraws with every data update and links back to this page.

Get an email when Reflection ships a new model

And once when it gets ranked, plus the weekly review on Mondays.

Choose providers and emails

Reasoning-effort configurations

The same model behaves differently depending on how much it is allowed to think. Each row is one setting, scored only on the benchmarks that were run at that setting. “Not stated” collects results from publishers that did not say which setting they used; for a reasoning model that is usually its thinking mode, but we do not assume it. Pick a setting here or at the top of the page to see its price, speed and category scores.

EffortIndexRankSpeedFirst answerChat reply costCategories
low––202 tok/s0.34 s$0.0005
medium––202 tok/s0.34 s$0.0005
xhigh––202 tok/s0.34 s$0.0005
not statedbest54.2#236202 tok/s0.34 s$0.0005Agents & tools 41 · Coding 52 · Human preference 61 · Maths 65 · Reasoning 62

How its index has moved

21 Oct 2026 to 15 Nov 2026
475563

The index is recomputed from scratch every day, so a line moves when a new benchmark result lands, when a publisher revises a score, or when the models it is normalised against change. Early movement usually means the score is still settling.

Benchmark results

See all LLM benchmarks. One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.

Reasoning

Benchmarknot statedSourceTrend
Arena Hard Prompts1461#96Arena (LMArena)

Coding

Benchmarklowmediumxhighnot statedSourceTrend
Arena Codingeffort not stated–––1505#69Arena (LMArena)
Arena WebDeveffort not stated–––1437#88Arena (LMArena)
GameBench 2 Overallnot in indexacross settings–––42.8#44DuelLab GameBench 2
GameBench 2 by reasoning settingnot in index39.1#14135.8#15253.4#86–DuelLab GameBench 2

Agents & tools

Benchmarknot statedSourceTrend
Arena Agent-5.9#44Arena (LMArena)

Maths

Benchmarknot statedSourceTrend
Arena Maths1471#52Arena (LMArena)

Instruction following

Benchmarknot statedSourceTrend
Arena Instruction Followingnot in index1433#94Arena (LMArena)

Human preference

Benchmarknot statedSourceTrend
Arena Text1434#107Arena (LMArena)
Arena Creative Writingnot in index1388#132Arena (LMArena)
Arena Multi-turnnot in index1432#114Arena (LMArena)
Arena Longer Queriesnot in index1438#115Arena (LMArena)

Safety & honesty

Benchmarknot statedSourceTrend
FORTRESSnot in index11.9%#77Scale AI SEAL

GameBench 2 Overall: 42.8. GameBench 2, best reasoning setting (XHigh): 53.4. Reference only, not part of the BenchLeader Index.

DuelLab GameBench 2, release gb2-20261111-7b598bb13de4 (https://benchmarks.duellab.org/releases/gb2-20261111-7b598bb13de4/), CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Scores mapped to BenchLeader model names; not endorsed by DuelLab.

API pricing: where to run it

Every provider serving this model through OpenRouter, with throughput and first-token latency measured on live traffic over the last 30 minutes and each provider’s own price. Purple marks the best in each column.

ProviderSpeedFirst tokenInput $/MOutput $/MContextQuantisation
Baseten432 tok/s0.47 s$0.400$1.201.0Mfp4
Nebius Token Factory192 tok/s0.83 s$0.250$1.001.0Mfp4
CoreWeave176 tok/s0.30 s$0.300$1.001.0Mnvfp4
Together113 tok/s0.34 s$0.400$1.201.0M–

Price history

The price above, per 1M tokens, every day it changed.

$0.00$0.825$1.65$2.48$3.30Oct 26
input output
Now $0.400 in · $1.20 out

What a task costs

Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache at $0.040 per 1M. Reasoning tokens are not modelled.

WorkloadTokens in / outCostWith cachingTime
Chat reply400 / 300$0.0005$0.00041.8 s
Summarize a 30-page report12,000 / 600$0.0055$0.00233.3 s
Code edit6,000 / 1,500$0.0042$0.00267.8 s
Agentic coding session60,000 / 4,000$0.029$0.01320.1 s
Structured extraction2,000 / 200$0.0010$0.00051.3 s

More from Reflection

New this month

Compare Beam 1.0 501B A23B with…

See also

Embed or cite

Show where Beam 1.0 501B A23B ranks on a model card or README. The badge redraws itself with every daily update, carries the data date and links back to this page.

[![Beam 1.0 501B A23B on BenchLeader: #113 overall](https://www.benchleader.com/badge/beam-1-0-501b-a23b.svg)](https://www.benchleader.com/models/beam-1-0-501b-a23b)

Cite this

The ranking changes daily, so a reference carries the data date. APA style:

BenchLeader. (2026, November 15). Beam 1.0 501B A23B on the BenchLeader Index (method v1.4) [Data set]. Retrieved November 15, 2026, from https://www.benchleader.com/models/beam-1-0-501b-a23b

BibTeX:

@misc{benchleader_beam_1_0_501b_a23b_20261115,
  author       = {{BenchLeader}},
  title        = {{Beam 1.0 501B A23B on the BenchLeader Index (method v1.4)}},
  year         = {2026},
  month        = nov,
  howpublished = {\url{https://www.benchleader.com/models/beam-1-0-501b-a23b}},
  note         = {Data as of 2026-11-15}
}

Every badge style, the embeddable top-10 widget and how to cite the Index

Questions people ask

What are Beam 1.0 501B A23B's benchmark scores?
As of 15 Nov 2026, Beam 1.0 501B A23B scores 54.2 ±7.6 on the BenchLeader Index, strongest in maths (65) and weakest in agents & tools (41). That rests on independent benchmark results only, and every score on this page links to its publisher.
How does Beam 1.0 501B A23B rank?
Beam 1.0 501B A23B is #236 of 611 ranked configurations on 15 Nov 2026. 380 other configurations score within its margin, so the exact position matters less than the band. The leader is Claude Fable 5.2 (max effort) at 72.7.
How much does Beam 1.0 501B A23B cost per 1M tokens?
As of 15 Nov 2026, Beam 1.0 501B A23B costs $0.400 per 1M input tokens and $1.20 per 1M output tokens, $0.600 blended (three input tokens per output token). The list price last changed on 19 Oct 2026.
Is Beam 1.0 501B A23B better than Muse Spark Open?
As of 15 Nov 2026, Beam 1.0 501B A23B and Muse Spark Open (xhigh) are neck and neck on quality (54.2 vs 53.5 on the BenchLeader Index, inside the ±10.0 margin of error), and Beam 1.0 501B A23B costs 3.3× less, which makes it the better buy for most workloads. The head-to-head page has the benchmark-by-benchmark detail.
When was Beam 1.0 501B A23B released?
Reflection released Beam 1.0 501B A23B on 5 Oct 2026. BenchLeader has results for it at 4 reasoning-effort settings.

Data as of 15 Nov 2026. Compare these configurations.

Cite as: BenchLeader, “Beam 1.0 501B A23B: benchmarks, pricing, speed and rank”, data as of 15 Nov 2026, https://www.benchleader.com/models/beam-1-0-501b-a23b.