Celeris-1 vs Llama 4 Maverick

Change models
Celeris-1Celeris
IndexAll categories, out of 100.
33.7 ±8.2
34.4 ±4.7 best
GapIndex points below the top.
−0.7 · no clear gap
Top
CoverageIndex weight that has results.
75%
95% best
SpeedOutput tokens per second.
722/s best
63/s
EloLMArena rating and rank.
—
—
ReleasedMonth of release.
2026-07
2025-04
Input / 1MUS dollars per 1M input tokens.
$0.20 best
$0.35
Output / 1MUS dollars per 1M output tokens.
$0.70 best
$1.00
1K chats2K tokens in, 500 out, each.
$0.75 best
$1.20
1K documents20K tokens in, 1K out, each.
$4.70 best
$8.00
1K agent tasks200K tokens in, 20K out, each.
$54.00 best
$90.00
ContextMaximum tokens in one request.
128K
524K best
AgenticMulti-step tasks with tools.
33.9
36.7 best
CodingCode writing and repair.
24.4
37.1 best
ReasoningLogic problems and puzzles.
35.0
36.7 best
MultimodalTasks with images and text.
—
35.9
KnowledgeFacts and expert knowledge.
35.3 best
33.9
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
26.2
29.6 best
MathMath problems.
—
28.5

Benchmarks

6 shared
Only Celeris-1: 4Only Llama 4 Maverick: 13
Celeris-1
Llama 4 Maverick
6.8% best158 / 216
4.9%174 / 216
38.0%163 / 211
50.0% best148 / 211
0.0%145 / 207
0.0%145 / 207
63.1%144 / 193
67.1% best133 / 193
0.0%100 / 132
0.0%100 / 132
21.6%111 / 113
31.7% best102 / 113