Muse Spark vs Beam 1.0 501B-A23B FP8
Change modelsMuse SparkMeta
Beam 1.0 501B-A23B FP8Reflection AI
IndexAll categories, out of 100.
59.8 ±3.5 best
59.2 ±5.4
GapIndex points below the top.
Top
−0.6 · no clear gap
CoverageIndex weight that has results.
95% best
85%
SpeedOutput tokens per second.
—
—
EloLMArena rating and rank.
1474
—
ReleasedMonth of release.
2026-04
2026-10
Input / 1MUS dollars per 1M input tokens.
—
—
Output / 1MUS dollars per 1M output tokens.
—
—
1K chats2K tokens in, 500 out, each.
—
—
1K documents20K tokens in, 1K out, each.
—
—
1K agent tasks200K tokens in, 20K out, each.
—
—
ContextMaximum tokens in one request.
262K
1M best
AgenticMulti-step tasks with tools.
55.9 best
55.4
CodingCode writing and repair.
53.3
61.4 best
ReasoningLogic problems and puzzles.
59.8
64.3 best
MultimodalTasks with images and text.
63.0
—
KnowledgeFacts and expert knowledge.
67.8 best
60.3
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
60.0
64.4 best
MathMath problems.
60.2 best
55.8
Benchmarks
7 sharedOnly Muse Spark: 40Only Beam 1.0 501B-A23B FP8: 11
Muse Spark
Beam 1.0 501B-A23B FP8