Claude 3 Haiku vs Phi-4
Change modelsClaude 3 HaikuAnthropic
Phi-4Microsoft
IndexAll categories, out of 100.
22.0 ±5.4
22.4 ±5.7 best
GapIndex points below the top.
−0.4 · no clear gap
Top
CoverageIndex weight that has results.
95% best
90%
SpeedOutput tokens per second.
73/s
93/s best
EloLMArena rating and rank.
1195
1217 best
ReleasedMonth of release.
2024-03
2024-12
Input / 1MUS dollars per 1M input tokens.
$0.25
$0.07 best
Output / 1MUS dollars per 1M output tokens.
$1.25
$0.14 best
1K chats2K tokens in, 500 out, each.
$1.13
$0.21 best
1K documents20K tokens in, 1K out, each.
$6.25
$1.54 best
1K agent tasks200K tokens in, 20K out, each.
$75.00
$16.80 best
ContextMaximum tokens in one request.
200K best
16K
AgenticMulti-step tasks with tools.
16.8 best
5.0
CodingCode writing and repair.
31.2 best
23.4
ReasoningLogic problems and puzzles.
31.6 best
22.1
MultimodalTasks with images and text.
5.0
—
KnowledgeFacts and expert knowledge.
24.1
29.9 best
MultilingualTasks in many languages.
—
24.0
InstructionTasks with strict prompt rules.
22.8 best
11.0
MathMath problems.
8.1
23.7 best
Benchmarks
9 sharedOnly Claude 3 Haiku: 3Only Phi-4: 5
Claude 3 Haiku
Phi-4