Gemma 4 31B vs GPT-5 mini

Change models
IndexAll categories, out of 100.
47.8 ±3.3
48.0 ±3.9 best
GapIndex points below the top.
−0.2 · no clear gap
Top
CoverageIndex weight that has results.
100% best
95%
SpeedOutput tokens per second.
36/s
77/s best
EloLMArena rating and rank.
1442 best
1374
ReleasedMonth of release.
2026-04
2025-08
Input / 1MUS dollars per 1M input tokens.
Free best
$0.25
Output / 1MUS dollars per 1M output tokens.
Free best
$2.00
Batch output / 1MBatch price per 1M output tokens.
—
$1.00
1K chats2K tokens in, 500 out, each.
Free best
$1.50
1K documents20K tokens in, 1K out, each.
Free best
$7.00
1K agent tasks200K tokens in, 20K out, each.
Free best
$90.00
ContextMaximum tokens in one request.
256K
400K best
AgenticMulti-step tasks with tools.
45.8 best
41.2
CodingCode writing and repair.
49.4 best
49.0
ReasoningLogic problems and puzzles.
46.1
46.6 best
MultimodalTasks with images and text.
56.4
57.2 best
KnowledgeFacts and expert knowledge.
44.7
45.9 best
MultilingualTasks in many languages.
47.4
51.6 best
InstructionTasks with strict prompt rules.
58.7
—
MathMath problems.
51.1
51.6 best

Benchmarks

15 shared
Only Gemma 4 31B: 32Only GPT-5 mini: 35
Gemma 4 31B
GPT-5 mini
EuroEval PolishMultilingual
61.6% best19 / 227
57.2%31 / 227
75.8% best106 / 220
75.0%110 / 220
EuroEval SwedishMultilingual
63.1%28 / 213
66.0% best19 / 213
EuroEval GermanMultilingual
55.7% best20 / 208
52.5%30 / 208
73.3%97 / 203
86.7% best72 / 203
67.5% best14 / 199
55.9%49 / 199
EuroEval FrenchMultilingual
67.9% best16 / 195
60.1%41 / 195
EuroEval SpanishMultilingual
55.8% best28 / 194
54.2%31 / 194
EuroEval DutchMultilingual
62.7% best20 / 192
60.1%27 / 192
EuroEval ItalianMultilingual
66.8% best14 / 191
60.1%31 / 191
61.4%60 / 97
66.9% best31 / 97
10.4%92 / 94
21.6% best79 / 94
39.3% best35 / 66
27.0%50 / 66
52.6%42 / 53
68.5% best4 / 53
50.8%27 / 51
51.9% best26 / 51