Claude 3 Haiku vs Gemma 4 E2B

Change models
IndexAll categories, out of 100.
22.0 ±5.4 best
21.9 ±5.5
GapIndex points below the top.
Top
−0.1 · no clear gap
CoverageIndex weight that has results.
95% best
90%
SpeedOutput tokens per second.
73/s
—
EloLMArena rating and rank.
1195
—
ReleasedMonth of release.
2024-03
2026-04
Input / 1MUS dollars per 1M input tokens.
$0.25
—
Output / 1MUS dollars per 1M output tokens.
$1.25
—
1K chats2K tokens in, 500 out, each.
$1.13
—
1K documents20K tokens in, 1K out, each.
$6.25
—
1K agent tasks200K tokens in, 20K out, each.
$75.00
—
ContextMaximum tokens in one request.
200K best
128K
AgenticMulti-step tasks with tools.
16.8
25.3 best
CodingCode writing and repair.
31.2 best
20.5
ReasoningLogic problems and puzzles.
31.6 best
25.6
MultimodalTasks with images and text.
5.0
16.8 best
KnowledgeFacts and expert knowledge.
24.1 best
22.1
MultilingualTasks in many languages.
—
5.0
InstructionTasks with strict prompt rules.
22.8
24.5 best
MathMath problems.
8.1
—

Benchmarks

7 shared
Only Claude 3 Haiku: 5Only Gemma 4 E2B: 15
Claude 3 Haiku
Gemma 4 E2B
4.1%199 / 222
4.8% best181 / 222
27.7% best184 / 217
16.3%195 / 217
0.0%150 / 214
0.0%150 / 214
37.4%192 / 199
43.3% best183 / 199
21.1% best129 / 149
20.8%130 / 149
36.1%121 / 146
38.0% best114 / 146
30.8%111 / 113
44.6% best107 / 113