Claude Opus 4.7 vs GLM-5.3-Flash

Change models
GLM-5.3-FlashZ.AIBudget
IndexAll categories, out of 100.
62.6 ±2.7 best
62.4 ±2.1
GapIndex points below the top.
Top
−0.2 · no clear gap
CoverageIndex weight that has results.
95%
100% best
SpeedOutput tokens per second.
46/s
157/s best
EloLMArena rating and rank.
1483 best
1471
ReleasedMonth of release.
2026-04
2026-08
Input / 1MUS dollars per 1M input tokens.
$5.00
$0.04 best
Output / 1MUS dollars per 1M output tokens.
$25.00
$0.078 best
Batch output / 1MBatch price per 1M output tokens.
$12.50
—
1K chats2K tokens in, 500 out, each.
$22.50
$0.119 best
1K documents20K tokens in, 1K out, each.
$125.00
$0.876 best
1K agent tasks200K tokens in, 20K out, each.
$1,500.00
$9.55 best
ContextMaximum tokens in one request.
1M
1.05M best
AgenticMulti-step tasks with tools.
61.3
68.7 best
CodingCode writing and repair.
63.8 best
62.2
ReasoningLogic problems and puzzles.
61.9 best
58.2
MultimodalTasks with images and text.
60.8
64.1 best
KnowledgeFacts and expert knowledge.
64.5 best
57.2
MultilingualTasks in many languages.
—
63.8
InstructionTasks with strict prompt rules.
52.2 best
34.2
MathMath problems.
63.2 best
56.6

Benchmarks

35 shared
Only Claude Opus 4.7: 50Only GLM-5.3-Flash: 57
Claude Opus 4.7
GLM-5.3-Flash
86.4%70 / 220
90.2% best43 / 220
86.7%72 / 203
93.9% best52 / 203
85.3% best22 / 149
83.9%45 / 149
TaxEval v2Finance
75.3%19 / 143
75.6% best14 / 143
85.1% best35 / 142
80.5%67 / 142
90.2% best25 / 135
86.4%43 / 135
MMLU ProKnowledge
89.9% best8 / 135
86.1%49 / 135
71.0% best54 / 130
30.8%89 / 130
MedScribeHealthcare
83.0%50 / 111
88.9% best10 / 111
38.5%44 / 98
45.2% best25 / 98
Chess PuzzlesReasoning
41.0% best37 / 93
21.0%62 / 93
70.2% best32 / 93
55.8%52 / 93
MMMU ProMultimodal
85.5%24 / 92
86.0% best23 / 92
23.7%51 / 89
31.6% best31 / 89
SWE-benchCoding
82.0%21 / 87
92.0% best10 / 87
28.0% best37 / 83
8.0%69 / 83
78.3% best25 / 80
76.4%35 / 80
87.2% best29 / 80
77.6%65 / 80
51.5%40 / 79
57.9% best13 / 79
43.9% best25 / 77
20.5%52 / 77