Apodex 1.1 vs Beam 1.0 501B-A23B NVFP4

Change models
Apodex 1.1Apodex
IndexAll categories, out of 100.
58.9 ±6.3 best
58.5 ±5.4
GapIndex points below the top.
Top
−0.4 · no clear gap
CoverageIndex weight that has results.
80%
85% best
SpeedOutput tokens per second.
—
—
EloLMArena rating and rank.
—
—
ReleasedMonth of release.
2026-08
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.
—
1M
AgenticMulti-step tasks with tools.
63.5 best
55.0
CodingCode writing and repair.
56.8
60.8 best
ReasoningLogic problems and puzzles.
53.7
62.7 best
MultimodalTasks with images and text.
58.2
—
KnowledgeFacts and expert knowledge.
57.4
59.2 best
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
—
63.0
MathMath problems.
—
55.5

Benchmarks

6 shared
Only Apodex 1.1: 12Only Beam 1.0 501B-A23B NVFP4: 12
Apodex 1.1
Beam 1.0 501B-A23B NVFP4
79.3% best52 / 217
78.7%61 / 217
4.6%87 / 214
15.4% best48 / 214
77.7%31 / 114
80.7% best15 / 114
70.8%50 / 86
78.4% best40 / 86
56.1% best16 / 36
51.1%25 / 36
92.4% best7 / 24
81.1%13 / 24