Beam 1.0 501B-A23B

  • Open weights
  • Reasoning
  • Input: text
  • Released 2026-10
RankedEnough results for an index.
IndexAll categories, out of 100.59.4±4.6Median 50.7
CoverageIndex weight that has results.85%
SpeedOutput tokens per second.259/sMedian 76/s
First tokenSeconds to the first output token.9.6 sMedian 1.1 s
Input / 1MUS dollars per 1M input tokens.—Median $0.60
Output / 1MUS dollars per 1M output tokens.—Median $2.50
ContextMaximum tokens in one request.1MMedian 262K
EloLMArena rating and rank.—Median 1425

Capabilities

Index per category, out of 100.
Out of 100
AgenticMulti-step tasks with tools.15 counted58.775 / 221
CodingCode writing and repair.3 counted61.161 / 214
ReasoningLogic problems and puzzles.4 counted62.257 / 225
MultimodalTasks with images and text.None counted——
KnowledgeFacts and expert knowledge.3 counted56.387 / 266
MultilingualTasks in many languages.None counted——
InstructionTasks with strict prompt rules.1 counted62.3—
MathMath problems.1 counted54.9—

Results

27 counted
BenchmarkResultThe score from the source.PlacePlace among results on this benchmark.
AIME 2026Math97.7%3 / 34
LongBench v2Reasoning67.5%3 / 25
SWE-bench ProCoding66.1%12 / 83
Critical Physics TasksReasoning16.3%46 / 207
Scientific Code BenchmarkCoding49.2%8 / 35
Instruction Following BenchmarkInstruction79.6%14 / 57
MCP AtlasAgentic80.8%15 / 53
GPQA Diamond · BenchLMKnowledge90.6%27 / 87
GDPval-AA normalizedAgentic33.9%54 / 132
Artificial Analysis CyberGym-E2E-AAAgenticAbove fitFar above this model's index.45.0%9 / 21
Artificial Analysis SciCodeCoding48.5%57 / 113
BrowseCompAgentic76.8%36 / 63
DeepSearchQAAgentic80.5%15 / 25
DeepSWEAgentic43.3%59 / 64
Artificial Analysis GDP.pdfAgentic11.6%22 / 23

Sources

BenchLMCC BY-NC 4.0

History

13 releases
Index59.9 → 59.4
23 Oct 202621 Nov 2026

Questions

How good is Beam 1.0 501B-A23B?
Its index is 59.4 out of 100. That is place 78 of 282 ranked models.
What is Beam 1.0 501B-A23B best at?
Reasoning: 62.2 out of 100, place 57 of 225.
How fast is Beam 1.0 501B-A23B?
259 output tokens per second, 9.6 s to the first token.
What is the context window of Beam 1.0 501B-A23B?
1M tokens.
Is Beam 1.0 501B-A23B open source?
Yes. Its weights are published.
When was Beam 1.0 501B-A23B released?
October 2026.

Badge

Beam 1.0 501B-A23B on Inference360: place 78 of 282, index 59.4

Compare with

Qwen3.7 Max59.5Sage-A1.559.5Claude Opus 4.6 (Adaptive)59.3Beam 1.0 501B-A23B FP858.4

More from Reflection AI

All models
Beam 1.0 501B-A23B FP858.4Beam 1.0 501B-A23B NVFP457.6