Grok 4.5 vs Grok 4.6

Change models
IndexAll categories, out of 100.
63.7 ±2.9
67.8 ±3.7 best
GapIndex points below the top.
−4.1 · no clear gap
Top
CoverageIndex weight that has results.
95% best
85%
SpeedOutput tokens per second.
53/s
56/s best
EloLMArena rating and rank.
1448 best
1430
ReleasedMonth of release.
2026-07
2026-08
Input / 1MUS dollars per 1M input tokens.
$2.00
$2.00
Output / 1MUS dollars per 1M output tokens.
$6.00
$6.00
Batch output / 1MBatch price per 1M output tokens.
—
$4.80
1K chats2K tokens in, 500 out, each.
$7.00
$7.00
1K documents20K tokens in, 1K out, each.
$46.00
$46.00
1K agent tasks200K tokens in, 20K out, each.
$520.00
$520.00
ContextMaximum tokens in one request.
500K
500K
AgenticMulti-step tasks with tools.
61.6
68.4 best
CodingCode writing and repair.
65.8
67.7 best
ReasoningLogic problems and puzzles.
63.3
66.1 best
MultimodalTasks with images and text.
47.7
—
KnowledgeFacts and expert knowledge.
65.8 best
65.7
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
71.3
72.0 best
MathMath problems.
61.8
65.3 best

Benchmarks

66 shared
Only Grok 4.5: 8Only Grok 4.6: 18
Grok 4.5
Grok 4.6
42.7%36 / 222
42.9% best34 / 222
79.3%52 / 217
80.3% best40 / 217
93.4% best18 / 217
93.2%21 / 217
15.4%48 / 214
17.1% best42 / 214
97.8%33 / 200
99.2% best19 / 200
93.1%12 / 199
94.9% best3 / 199
51.6% best16 / 164
48.2%27 / 164
86.0%17 / 149
86.3% best14 / 149
TaxEval v2Finance
71.7% best74 / 143
71.1%83 / 143
87.4%15 / 142
88.2% best7 / 142
92.9%12 / 135
94.7% best3 / 135
MMLU ProKnowledge
89.2%14 / 135
89.4% best11 / 135
44.2%41 / 133
55.9% best16 / 133
52.6%50 / 131
67.1% best31 / 131
CorpFin v2Finance
67.4% best12 / 131
66.2%20 / 131
85.7%50 / 126
87.0% best45 / 126
69.0%51 / 123
76.2% best41 / 123
58.2%43 / 122
68.3% best28 / 122
94.8%35 / 116
96.3% best26 / 116
55.0%29 / 114
56.5% best24 / 114