Hy4 preview vs Sakana Fugu-Ultra

Change models
Hy4 previewTencent
IndexAll categories, out of 100.
66.4 ±4.1 best
66.3 ±7.7
GapIndex points below the top.
Top
−0.1 · no clear gap
CoverageIndex weight that has results.
90% best
80%
SpeedOutput tokens per second.
67/s
—
EloLMArena rating and rank.
—
—
ReleasedMonth of release.
2026-08
2026-06
Input / 1MUS dollars per 1M input tokens.
$0.834
—
Output / 1MUS dollars per 1M output tokens.
$2.50
—
1K chats2K tokens in, 500 out, each.
$2.92
—
1K documents20K tokens in, 1K out, each.
$19.18
—
1K agent tasks200K tokens in, 20K out, each.
$216.82
—
ContextMaximum tokens in one request.
1.05M best
1M
AgenticMulti-step tasks with tools.
63.5
69.2 best
CodingCode writing and repair.
65.5
68.3 best
ReasoningLogic problems and puzzles.
69.6 best
68.4
MultimodalTasks with images and text.
75.7 best
62.8
KnowledgeFacts and expert knowledge.
67.9
69.7 best
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
—
—
MathMath problems.
73.0
—

Benchmarks

5 shared
Only Hy4 preview: 48Only Sakana Fugu-Ultra: 5
Hy4 preview
Sakana Fugu-Ultra
92.3%18 / 141
95.5% best2 / 141
85.4% best18 / 86
82.1%28 / 86
92.3%19 / 84
95.5% best2 / 84
65.7%14 / 83
73.7% best8 / 83
43.4%20 / 53
50.0% best9 / 53