Reflection: Beam 1.0 501B A23B

Beam 1.0 is Reflection's open-weight reasoning model for coding and long-horizon agentic workflows. It uses a sparse mixture-of-experts architecture with 501B total and 23B active parameters per toke…

reflection/beam-1.0-501b-a23b-20261016

Beam 1.0 is Reflection's open-weight reasoning model for coding and long-horizon agentic workflows. It uses a sparse mixture-of-experts architecture with 501B total and 23B active parameters per token, and a 1M-token context window.

Beam targets software engineering, multi-step tool use, and mathematical and scientific reasoning, with an emphasis on token efficiency. It takes text input and produces text output, supports configurable reasoning effort, function calling and structured outputs, and is released under Apache 2.0.

Specifications & capabilities

Input
text
Output
text
Context window · tokens
1048576
Knowledge cutoff
2026-06-30
Published model ID
Beam-1.0-501B-A23B
Capabilities
promptCaching, reasoning, structuredOutputs, toolCalling
Description source ↗

Evaluations

21–40 / 83

LMArena

1,489.823

source id
lmarena
leaderboard name
LMArena Text Factuality Hard Prompts English
dimension
factuality
metric
Arena rating
rank
59
sample size
935
evaluated at
2026-11-09T00:00:00.000Z
source model id
Beam 1.0 501B A23B
confidence lower
1473.3108004105184
confidence upper
1506.3350421493562
Original result ↗

LMArena

1,459.396

source id
lmarena
leaderboard name
LMArena Text Math
dimension
general
metric
Arena rating
rank
59
sample size
207
evaluated at
2026-11-09T00:00:00.000Z
source model id
Beam 1.0 501B A23B
confidence lower
1420.5770995367056
confidence upper
1498.2157158833122
Original result ↗

LMArena

1,474.604

source id
lmarena
leaderboard name
LMArena Text Coding
dimension
general
metric
Arena rating
rank
62
sample size
961
evaluated at
2026-11-09T00:00:00.000Z
source model id
Beam 1.0 501B A23B
confidence lower
1455.9226987841291
confidence upper
1493.2858892904505
Original result ↗

LMArena

1,461.127

source id
lmarena
leaderboard name
LMArena Text French
dimension
general
metric
Arena rating
rank
62
sample size
244
evaluated at
2026-11-09T00:00:00.000Z
source model id
Beam 1.0 501B A23B
confidence lower
1421.1986968915955
confidence upper
1501.0549628782323
Original result ↗

LMArena

1,450.62

source id
lmarena
leaderboard name
LMArena Text Factuality Instruction Following
dimension
factuality
metric
Arena rating
rank
64
sample size
1438
evaluated at
2026-11-09T00:00:00.000Z
source model id
Beam 1.0 501B A23B
confidence lower
1437.5281970163117
confidence upper
1463.710869813009
Original result ↗

LMArena

1,504.66

source id
lmarena
leaderboard name
LMArena Text Style Control Coding
dimension
style
metric
Arena rating
rank
70
sample size
961
evaluated at
2026-11-09T00:00:00.000Z
source model id
Beam 1.0 501B A23B
confidence lower
1485.7018711393757
confidence upper
1523.6190576669092
Original result ↗

LMArena

1,462.506

source id
lmarena
leaderboard name
LMArena Text Expert
dimension
general
metric
Arena rating
rank
73
sample size
362
evaluated at
2026-11-09T00:00:00.000Z
source model id
Beam 1.0 501B A23B
confidence lower
1432.7968628177941
confidence upper
1492.2144741069674
Original result ↗

LMArena

1,470.972

source id
lmarena
leaderboard name
LMArena Text Factuality Hard Prompts
dimension
factuality
metric
Arena rating
rank
74
sample size
2513
evaluated at
2026-11-09T00:00:00.000Z
source model id
Beam 1.0 501B A23B
confidence lower
1460.6333903314166
confidence upper
1481.3115425925153
Original result ↗

LMArena

1,480.071

source id
lmarena
leaderboard name
LMArena Text Style Control Expert
dimension
style
metric
Arena rating
rank
78
sample size
362
evaluated at
2026-11-09T00:00:00.000Z
source model id
Beam 1.0 501B A23B
confidence lower
1450.4629308776016
confidence upper
1509.6795954511258
Original result ↗

LMArena

1,454.035

source id
lmarena
leaderboard name
LMArena Text Hard Prompts English
dimension
general
metric
Arena rating
rank
78
sample size
935
evaluated at
2026-11-09T00:00:00.000Z
source model id
Beam 1.0 501B A23B
confidence lower
1434.7453951426412
confidence upper
1473.3243580987157
Original result ↗

OpenRouter Benchmarks

69.333%

source id
openRouterBenchmarks
leaderboard name
tau2-bench Verified Airline · Accuracy · auto-routing
dimension
agentic
metric
tau2-bench Verified Airline · Accuracy
rank
78
sample size
50
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
1
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
tau_bench_verified_airline
name
tau2-bench Verified Airline
record_kind
aggregate
provider_name
auto-routing
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-24T12:03:24.771Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
tau2-bench Verified Airline · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
50
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/tau2-bench-airline
measurements
accuracy
0.693333
cost_per_task_usd
0.02754302
output_tokens_per_task
14914.54
latency_seconds_per_task
209.40879999999999
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
128
benchmark_evaluation
true
openrouter_permaslug
reflection/beam-1.0-501b-a23b-20261016
source model id
reflection/beam-1.0-501b-a23b-20261016
Original result ↗

LMArena

1,458.317

source id
lmarena
leaderboard name
LMArena Text Factuality English
dimension
factuality
metric
Arena rating
rank
79
sample size
1434
evaluated at
2026-11-09T00:00:00.000Z
source model id
Beam 1.0 501B A23B
confidence lower
1445.4094952708404
confidence upper
1471.2239409157846
Original result ↗

LMArena

1,461.592

source id
lmarena
leaderboard name
LMArena Text Industry Software And It Services
dimension
general
metric
Arena rating
rank
80
sample size
1369
evaluated at
2026-11-09T00:00:00.000Z
source model id
Beam 1.0 501B A23B
confidence lower
1445.752759750991
confidence upper
1477.4311139218955
Original result ↗

LMArena

1,417.475

source id
lmarena
leaderboard name
LMArena Text Factuality Industry Entertainment And Sports And Media
dimension
factuality
metric
Arena rating
rank
80
sample size
1149
evaluated at
2026-11-09T00:00:00.000Z
source model id
Beam 1.0 501B A23B
confidence lower
1402.5209449164956
confidence upper
1432.4298122493065
Original result ↗

LMArena

1,425.1

source id
lmarena
leaderboard name
LMArena Text Instruction Following
dimension
general
metric
Arena rating
rank
82
sample size
1438
evaluated at
2026-11-09T00:00:00.000Z
source model id
Beam 1.0 501B A23B
confidence lower
1409.2779711157405
confidence upper
1440.9214850357855
Original result ↗

LMArena

1,467.485

source id
lmarena
leaderboard name
LMArena Text Style Control French
dimension
style
metric
Arena rating
rank
83
sample size
244
evaluated at
2026-11-09T00:00:00.000Z
source model id
Beam 1.0 501B A23B
confidence lower
1427.2256593642933
confidence upper
1507.7441156145208
Original result ↗

LMArena

1,442.388

source id
lmarena
leaderboard name
LMArena Webdev Webdev React
dimension
coding
metric
Arena rating
rank
83
sample size
1584
evaluated at
2026-11-10T00:00:00.000Z
source model id
Beam 1.0 501B A23B
confidence lower
1424.8589561163742
confidence upper
1459.9164774996416
Original result ↗

LMArena

1,454.186

source id
lmarena
leaderboard name
LMArena Text Factuality Multi Turn
dimension
factuality
metric
Arena rating
rank
84
sample size
634
evaluated at
2026-11-09T00:00:00.000Z
source model id
Beam 1.0 501B A23B
confidence lower
1433.3340872302126
confidence upper
1475.037369603003
Original result ↗

LMArena

1,484.549

source id
lmarena
leaderboard name
LMArena Text Style Control Industry Software And It Services
dimension
style
metric
Arena rating
rank
87
sample size
1369
evaluated at
2026-11-09T00:00:00.000Z
source model id
Beam 1.0 501B A23B
confidence lower
1468.492147033438
confidence upper
1500.6062211867925
Original result ↗

LMArena

1,472.902

source id
lmarena
leaderboard name
LMArena Text Style Control Hard Prompts English
dimension
style
metric
Arena rating
rank
87
sample size
935
evaluated at
2026-11-09T00:00:00.000Z
source model id
Beam 1.0 501B A23B
confidence lower
1453.3720226894818
confidence upper
1492.4318324296084
Original result ↗

Published prices

1–3 / 3

Provider specifications & access

1–2 / 2

reflection/Beam-1.0-501B-A23B ↗

source model id
azure-huggingface/reflection--beam-1.0-501b-a23b@1
hosting
foundry-managed-compute
platforms
  • Microsoft Foundry
license
apache-2.0
context length
1048576
architecture
input_modalities
  • text
output_modalities
  • text
capabilities

    reflection/Beam-1.0-501B-A23B ↗

    source model id
    azure-huggingface/reflection--beam-1.0-501b-a23b@2
    hosting
    foundry-managed-compute
    platforms
    • Microsoft Foundry
    license
    apache-2.0
    context length
    1048576
    architecture
    input_modalities
    • text
    output_modalities
    • text
    capabilities

      Source specification observations

      1–5 / 5

      azureFoundry ↗

      field
      context_length
      value
      1048576
      normalized value
      1048576
      scope
      source-model
      source id
      azureFoundry
      confidence
      1

      azureFoundry ↗

      field
      context_length
      value
      1048576
      normalized value
      1048576
      scope
      source-model
      source id
      azureFoundry
      confidence
      1

      openRouter ↗

      field
      context_length
      value
      1048576
      normalized value
      1048576
      scope
      source-model
      source id
      openRouter
      confidence
      1

      openRouter ↗

      field
      knowledge_cutoff
      reported field
      knowledge_cutoff
      value
      2026-06-30
      normalized value
      2026-06-30
      precision
      day
      scope
      source-model
      source id
      openRouter
      confidence
      1

      openRouter ↗

      field
      max_output_tokens
      value
      943718
      normalized value
      943718
      scope
      source-model
      source id
      openRouter
      confidence
      1

      Original sources

      1–10 / 10

      Hugging Face · reflection/Beam-1.0-501B-A23B ↗

      Microsoft Foundry · Model page ↗

      Microsoft Foundry · Model page ↗

      OpenRouter · Model page ↗

      Microsoft Foundry · Model documentation ↗

      Hugging Face · Catalog ↗

      LMArena · Catalog ↗

      Microsoft Foundry · Catalog ↗

      OpenRouter · Catalog ↗

      OpenRouter · Catalog ↗

      Source links

      1–7 / 7

      Explore more models

      Models

      Loading
      EVIDENCE / COMPARISON

      Benchmarks

      Explore every benchmark

      Loading

      These charts contain model-linked results. The charts above also include unlinked source profiles. Compare scores within each evaluation. Model quality, usage, transparency and system performance measure different things.

      Install TensorTrace