Mercury 2.5 vs Mercury 3
Change modelsMercury 2.5Inception
Mercury 3Inception
IndexAll categories, out of 100.
46.2 ±5.4
52.3 ±8.2 best
GapIndex points below the top.
−6.1 · no clear gap
Top
CoverageIndex weight that has results.
85% best
75%
SpeedOutput tokens per second.
125/s
295/s best
EloLMArena rating and rank.
—
—
ReleasedMonth of release.
2026-09
2026-10
Input / 1MUS dollars per 1M input tokens.
$0.04 best
$0.10
Output / 1MUS dollars per 1M output tokens.
$0.15 best
$0.35
1K chats2K tokens in, 500 out, each.
$0.155 best
$0.375
1K documents20K tokens in, 1K out, each.
$0.95 best
$2.35
1K agent tasks200K tokens in, 20K out, each.
$11.00 best
$27.00
ContextMaximum tokens in one request.
260K
1M best
AgenticMulti-step tasks with tools.
49.3 best
44.4
CodingCode writing and repair.
44.5
64.2 best
ReasoningLogic problems and puzzles.
45.4
57.0 best
MultimodalTasks with images and text.
—
—
KnowledgeFacts and expert knowledge.
41.4
59.0 best
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
61.0 best
60.8
MathMath problems.
48.2
—
Benchmarks
8 sharedOnly Mercury 2.5: 20Only Mercury 3: 5
Mercury 2.5
Mercury 3