Claude Fable 5 vs GPT-5.6 Terra

Change models
IndexAll categories, out of 100.
74.0 ±2.7 best
68.9 ±2.0
GapIndex points below the top.
Top
−5.1
CoverageIndex weight that has results.
95%
100% best
SpeedOutput tokens per second.
56/s
66/s best
EloLMArena rating and rank.
1492 best
1449
ReleasedMonth of release.
2026-06
2026-07
Input / 1MUS dollars per 1M input tokens.
$10.00
$2.00 best
Output / 1MUS dollars per 1M output tokens.
$50.00
$12.00 best
Batch output / 1MBatch price per 1M output tokens.
$25.00
$6.00 best
1K chats2K tokens in, 500 out, each.
$45.00
$10.00 best
1K documents20K tokens in, 1K out, each.
$250.00
$52.00 best
1K agent tasks200K tokens in, 20K out, each.
$3,000.00
$640.00 best
ContextMaximum tokens in one request.
1M
1.05M best
AgenticMulti-step tasks with tools.
71.4 best
68.2
CodingCode writing and repair.
74.7 best
68.4
ReasoningLogic problems and puzzles.
73.6 best
72.5
MultimodalTasks with images and text.
70.1 best
67.3
KnowledgeFacts and expert knowledge.
75.8 best
65.5
MultilingualTasks in many languages.
—
60.7
InstructionTasks with strict prompt rules.
64.1 best
56.4
MathMath problems.
76.2 best
72.8

Benchmarks

73 shared
Only Claude Fable 5: 19Only GPT-5.6 Terra: 36
Claude Fable 5
GPT-5.6 Terra
55.5% best5 / 222
42.9%34 / 222
82.3%27 / 217
83.0% best18 / 217
85.9%69 / 217
93.3% best19 / 217
28.6%14 / 214
30.0% best10 / 214
99.7%14 / 200
99.7%14 / 200
92.6% best17 / 199
92.5%18 / 199
65.4% best4 / 164
46.8%31 / 164
88.6% best1 / 149
85.1%25 / 149
98.5% best3 / 149
86.3%51 / 149
63.5%58 / 146
71.2% best39 / 146
TaxEval v2Finance
76.9% best5 / 143
76.2%6 / 143
89.8% best2 / 142
85.9%25 / 142
93.2% best10 / 135
90.9%24 / 135
MMLU ProKnowledge
91.5% best3 / 135
86.7%39 / 135
55.5% best18 / 133
47.6%33 / 133
89.2% best11 / 131
83.9%17 / 131
CorpFin v2Finance
71.8% best2 / 131
65.3%28 / 131
98.5% best1 / 126
96.5%11 / 126
90.4% best4 / 123
74.6%44 / 123
96.4% best9 / 122
91.3%13 / 122