Inference
EN

Gemini 4 Hydrogen vs MiniMax M3.1 Flash

Change models
Gemini 4 HydrogenGoogle
MiniMax M3.1 FlashMiniMax
IndexAll categories, out of 100.
77.3 ±8.0 best
69.1 ±5.4
GapIndex points below the top.
Top
−8.2 · no clear gap
CoverageIndex weight that has results.
65%
80% best
SpeedOutput tokens per second.
—
100/s
EloLMArena rating and rank.
—
—
ReleasedMonth of release.
2026-11
2026-10
Input / 1MUS dollars per 1M input tokens.
$1.50
$0.15 best
Output / 1MUS dollars per 1M output tokens.
$7.50
$0.60 best
1K chats2K tokens in, 500 out, each.
$6.75
$0.60 best
1K documents20K tokens in, 1K out, each.
$37.50
$3.60 best
1K agent tasks200K tokens in, 20K out, each.
$450.00
$42.00 best
ContextMaximum tokens in one request.
1M
1.05M best
AgenticMulti-step tasks with tools.
78.7 best
71.7
CodingCode writing and repair.
77.4 best
69.2
ReasoningLogic problems and puzzles.
—
67.7
MultimodalTasks with images and text.
80.6 best
58.0
KnowledgeFacts and expert knowledge.
74.3 best
68.7
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
—
—
MathMath problems.
—
—

Benchmarks

7 shared
Only Gemini 4 Hydrogen: 9Only MiniMax M3.1 Flash: 29
Gemini 4 Hydrogen
MiniMax M3.1 Flash
Artificial Analysis Omniscience AccuracyKnowledge
50.5% best19 / 165
45.1%38 / 165
Vibe Code Bench v1.1Coding
87.8% best12 / 124
73.9%46 / 124
Legal Research BenchLegal
53.8% best6 / 91
34.1%51 / 91
Tax Agent BenchFinance
52.3% best1 / 83
32.1%24 / 83
Terminal-Bench 4.0Agentic
54.0% best10 / 62
25.3%29 / 62
HLAB-Vals-1.1Legal
11.3% best19 / 36
7.5%25 / 36
Multimodal Knowledge Agent BenchmarkAgentic
53.4% best3 / 20
33.5%19 / 20