GPT-5.6 Luna vs Hy4 flash

Change models
Hy4 flashTencent
IndexAll categories, out of 100.
64.3 ±2.1
64.5 ±8.4 best
GapIndex points below the top.
−0.2 · no clear gap
Top
CoverageIndex weight that has results.
100% best
65%
SpeedOutput tokens per second.
70/s best
69/s
EloLMArena rating and rank.
1431
—
ReleasedMonth of release.
2026-07
2026-11
Input / 1MUS dollars per 1M input tokens.
$0.20
$0.035 best
Output / 1MUS dollars per 1M output tokens.
$1.20
$0.14 best
Batch output / 1MBatch price per 1M output tokens.
$0.60
—
1K chats2K tokens in, 500 out, each.
$1.00
$0.14 best
1K documents20K tokens in, 1K out, each.
$5.20
$0.841 best
1K agent tasks200K tokens in, 20K out, each.
$64.00
$9.81 best
ContextMaximum tokens in one request.
1.05M
1.05M
AgenticMulti-step tasks with tools.
63.6
67.2 best
CodingCode writing and repair.
66.2 best
59.6
ReasoningLogic problems and puzzles.
65.1
—
MultimodalTasks with images and text.
63.6 best
53.3
KnowledgeFacts and expert knowledge.
61.3
78.0 best
MultilingualTasks in many languages.
53.1
—
InstructionTasks with strict prompt rules.
48.2
—
MathMath problems.
67.2
—

Benchmarks

7 shared
Only GPT-5.6 Luna: 96Only Hy4 flash: 8
GPT-5.6 Luna
Hy4 flash
77.1% best45 / 130
74.8%49 / 130
36.5%53 / 98
38.5% best44 / 98
24.8% best49 / 89
23.3%52 / 89
11.6%49 / 68
15.2% best44 / 68
DeepSWEAgentic
67.2% best32 / 64
64.6%37 / 64
78.4% best24 / 58
75.8%33 / 58
17.6% best36 / 52
11.8%44 / 52