MiniMax M2.7 vs Nemotron 3.5 Super VL 120B A12B

Change models
IndexAll categories, out of 100.
51.2 ±4.5 best
51.0 ±5.1
GapIndex points below the top.
Top
−0.2 · no clear gap
CoverageIndex weight that has results.
85%
85%
SpeedOutput tokens per second.
35/s
—
EloLMArena rating and rank.
1405
—
ReleasedMonth of release.
2026-03
2026-10
Input / 1MUS dollars per 1M input tokens.
$0.21
—
Output / 1MUS dollars per 1M output tokens.
$0.84
—
1K chats2K tokens in, 500 out, each.
$0.84
—
1K documents20K tokens in, 1K out, each.
$5.04
—
1K agent tasks200K tokens in, 20K out, each.
$58.80
—
ContextMaximum tokens in one request.
197K
1M best
AgenticMulti-step tasks with tools.
52.0 best
47.8
CodingCode writing and repair.
52.3
53.0 best
ReasoningLogic problems and puzzles.
49.5
56.1 best
MultimodalTasks with images and text.
—
47.4
KnowledgeFacts and expert knowledge.
50.2
52.7 best
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
59.8 best
55.2
MathMath problems.
47.7
—

Benchmarks

13 shared
Only MiniMax M2.7: 40Only Nemotron 3.5 Super VL 120B A12B: 15
MiniMax M2.7
Nemotron 3.5 Super VL 120B A12B
29.6%82 / 222
30.7% best74 / 222
78.3% best63 / 217
68.6%121 / 217
0.6%132 / 214
10.9% best62 / 214
26.8% best92 / 164
26.2%97 / 164
25.5%72 / 133
33.1% best55 / 133
11.9% best103 / 123
3.7%111 / 123
50.1% best52 / 114
47.8%61 / 114
10.6% best79 / 90
10.1%81 / 90
87.0%49 / 84
89.5% best33 / 84
56.2% best50 / 83
51.8%65 / 83
6.9%75 / 82
11.6% best65 / 82
28.0%53 / 77
29.2% best49 / 77
67.6% best10 / 32
18.9%26 / 32