Mistral Medium 3.5 128B vs Qwen3.5-35B-A3B

Change models
IndexAll categories, out of 100.
49.7 ±5.3 best
49.2 ±3.2
GapIndex points below the top.
Top
−0.5 · no clear gap
CoverageIndex weight that has results.
90%
100% best
SpeedOutput tokens per second.
—
78/s
EloLMArena rating and rank.
—
1395
ReleasedMonth of release.
2026-04
2026-03
Input / 1MUS dollars per 1M input tokens.
$1.50
$0.14 best
Output / 1MUS dollars per 1M output tokens.
$7.50
$1.00 best
1K chats2K tokens in, 500 out, each.
$6.75
$0.78 best
1K documents20K tokens in, 1K out, each.
$37.50
$3.80 best
1K agent tasks200K tokens in, 20K out, each.
$450.00
$48.00 best
ContextMaximum tokens in one request.
256K
262K best
AgenticMulti-step tasks with tools.
59.4 best
48.2
CodingCode writing and repair.
52.8 best
48.3
ReasoningLogic problems and puzzles.
45.4
48.0 best
MultimodalTasks with images and text.
39.3
54.2 best
KnowledgeFacts and expert knowledge.
41.6
47.5 best
MultilingualTasks in many languages.
45.4
50.1 best
InstructionTasks with strict prompt rules.
52.6 best
51.8
MathMath problems.
—
49.2

Benchmarks

10 shared
Only Mistral Medium 3.5 128B: 11Only Qwen3.5-35B-A3B: 21
Mistral Medium 3.5 128B
Qwen3.5-35B-A3B
13.8%125 / 216
21.0% best107 / 216
69.3%110 / 211
72.0% best100 / 211
0.0%145 / 207
0.9% best119 / 207
74.8%113 / 193
84.5% best77 / 193
23.0% best117 / 170
20.1%132 / 170
94.2% best24 / 143
89.2%42 / 143
68.8%46 / 140
72.5% best34 / 140
20.7% best112 / 120
20.1%115 / 120
77.6% best34 / 114
69.2%76 / 114
64.9%80 / 109
72.7% best65 / 109