Inference
EN

Gemini 3.1 Flash Lite Preview vs Grok 4

Change models
Gemini 3.1 Flash Lite PreviewGoogle
Grok 4xAI
IndexAll categories, out of 100.
49.6 ±6.9
50.0 ±4.0 best
GapIndex points below the top.
−0.4 · no clear gap
Top
CoverageIndex weight that has results.
80%
95% best
SpeedOutput tokens per second.
151/s best
78/s
EloLMArena rating and rank.
1416 best
1411
ReleasedMonth of release.
2026-03
2025-07
Input / 1MUS dollars per 1M input tokens.
$0.25
—
Output / 1MUS dollars per 1M output tokens.
$1.50
—
1K chats2K tokens in, 500 out, each.
$1.25
—
1K documents20K tokens in, 1K out, each.
$6.50
—
1K agent tasks200K tokens in, 20K out, each.
$80.00
—
ContextMaximum tokens in one request.
1.05M best
128K
AgenticMulti-step tasks with tools.
41.3
48.0 best
CodingCode writing and repair.
48.3
50.9 best
ReasoningLogic problems and puzzles.
—
47.3
MultimodalTasks with images and text.
62.5 best
49.6
KnowledgeFacts and expert knowledge.
52.3
55.8 best
MultilingualTasks in many languages.
57.3
—
InstructionTasks with strict prompt rules.
—
39.2
MathMath problems.
51.9 best
50.5

Benchmarks

15 shared
Only Gemini 3.1 Flash Lite Preview: 16Only Grok 4: 25
Gemini 3.1 Flash Lite Preview
Grok 4
LegalBenchLegal
83.8% best47 / 149
83.2%57 / 149
TaxEval v2Finance
71.8% best72 / 143
65.1%116 / 143
LiveCodeBenchCoding
80.1%70 / 142
83.2% best49 / 142
GPQA Diamond · Vals AIKnowledge
81.1%64 / 135
88.1% best35 / 135
MMLU ProKnowledge
86.2% best45 / 135
85.3%58 / 135
CorpFin v2Finance
59.4%78 / 131
66.0% best23 / 131
MedScribeHealthcare
63.9%102 / 109
78.2% best66 / 109
MedCodeHealthcare
47.6% best32 / 106
38.1%72 / 106
MortgageTaxFinance
68.0% best20 / 97
44.5%87 / 97
AIMEMath
83.3%43 / 95
90.6% best27 / 95
MMMU ProMultimodal
82.5% best37 / 92
76.3%54 / 92
SWE-benchCoding
62.8% best74 / 87
57.8%78 / 87
Terminal-Bench 2.0 · Vals AIAgentic
24.7%52 / 66
28.1% best47 / 66
CaseLaw v2Legal
55.0%35 / 53
65.8% best9 / 53
Finance Agent v1.1Finance
46.1%33 / 51
53.5% best19 / 51