K-EXAONE 2.0 vs Qwen3.5-35B-A3B

Change models
K-EXAONE 2.0LG AI Research
IndexAll categories, out of 100.
48.4 ±5.4
48.9 ±3.2 best
GapIndex points below the top.
−0.5 · no clear gap
Top
CoverageIndex weight that has results.
85%
100% best
SpeedOutput tokens per second.
—
78/s
EloLMArena rating and rank.
—
1395
ReleasedMonth of release.
2026-07
2026-03
Input / 1MUS dollars per 1M input tokens.
—
$0.14
Output / 1MUS dollars per 1M output tokens.
—
$1.00
1K chats2K tokens in, 500 out, each.
—
$0.78
1K documents20K tokens in, 1K out, each.
—
$3.80
1K agent tasks200K tokens in, 20K out, each.
—
$48.00
ContextMaximum tokens in one request.
262K
262K
AgenticMulti-step tasks with tools.
55.5 best
46.8
CodingCode writing and repair.
49.1 best
48.4
ReasoningLogic problems and puzzles.
41.9
48.0 best
MultimodalTasks with images and text.
—
54.3
KnowledgeFacts and expert knowledge.
43.8
47.4 best
MultilingualTasks in many languages.
—
50.0
InstructionTasks with strict prompt rules.
52.4 best
51.8
MathMath problems.
50.3 best
49.1

Benchmarks

8 shared
Only K-EXAONE 2.0: 13Only Qwen3.5-35B-A3B: 23
K-EXAONE 2.0
Qwen3.5-35B-A3B
18.6%116 / 216
21.0% best107 / 216
56.2%139 / 211
72.0% best100 / 211
0.9%119 / 207
0.9%119 / 207
82.9%85 / 193
84.5% best77 / 193
14.5%160 / 170
20.1% best132 / 170
68.2%77 / 114
69.2% best76 / 114
83.5%27 / 73
85.3% best16 / 73
92.4% best18 / 69
91.9%21 / 69