Inference
EN

GPT-5.6 Sol vs Step 5 Preview

Change models
GPT-5.6 SolOpenAI
Step 5 PreviewStepFun
IndexAll categories, out of 100.
73.0 ±1.9 best
68.4 ±3.7
GapIndex points below the top.
Top
−4.6
CoverageIndex weight that has results.
100% best
90%
SpeedOutput tokens per second.
64/s
105/s best
EloLMArena rating and rank.
1457 best
1443
ReleasedMonth of release.
2026-07
2026-09
Input / 1MUS dollars per 1M input tokens.
$4.40
$1.00 best
Output / 1MUS dollars per 1M output tokens.
$22.00
$2.70 best
Batch output / 1MBatch price per 1M output tokens.
$10.00
—
1K chats2K tokens in, 500 out, each.
$19.80
$3.35 best
1K documents20K tokens in, 1K out, each.
$110.00
$22.70 best
1K agent tasks200K tokens in, 20K out, each.
$1,320.00
$254.00 best
ContextMaximum tokens in one request.
1.05M best
1M
AgenticMulti-step tasks with tools.
71.9 best
68.8
CodingCode writing and repair.
70.4
71.4 best
ReasoningLogic problems and puzzles.
76.3 best
67.4
MultimodalTasks with images and text.
72.6 best
62.2
KnowledgeFacts and expert knowledge.
72.7 best
63.5
MultilingualTasks in many languages.
66.4
—
InstructionTasks with strict prompt rules.
64.5
—
MathMath problems.
75.1 best
61.7

Benchmarks

39 shared
Only GPT-5.6 Sol: 86Only Step 5 Preview: 23
GPT-5.6 Sol
Step 5 Preview
Artificial Analysis Humanity's Last ExamKnowledge
49.5% best13 / 222
46.5%23 / 222
Artificial Analysis Long Context ReasoningReasoning
84.0%8 / 217
88.3% best2 / 217
Critical Physics TasksReasoning
32.3% best3 / 214
20.9%26 / 214
Artificial Analysis Omniscience AccuracyKnowledge
59.4% best8 / 164
36.4%64 / 164
GDPval-AA normalizedAgentic
55.4% best19 / 133
54.3%24 / 133
Vibe Code Bench v1.1Coding
80.5% best31 / 123
70.4%48 / 123
Artificial Analysis SciCodeCoding
57.1%22 / 114
58.9% best13 / 114
Artificial Analysis MMMU-ProMultimodal
83.4% best10 / 113
76.4%44 / 113
Legal Research BenchLegal
48.1% best12 / 90
40.4%35 / 90
Terminal-Bench 2.1 · BenchLMAgentic
91.9% best2 / 86
85.0%20 / 86
GPQA Diamond · BenchLMKnowledge
94.6% best4 / 84
93.5%10 / 84
Tax Agent BenchFinance
31.1% best29 / 82
27.3%39 / 82
LiveBench Data AnalysisReasoning
79.8% best12 / 80
63.5%69 / 80
LiveBench ReasoningReasoning
91.7% best5 / 80
83.8%44 / 80
Finance Agent (v2)Finance
53.8% best31 / 79
50.7%42 / 79
Code MigrationCoding
52.9% best12 / 77
39.0%31 / 77
EMBFinance
72.3% best7 / 75
57.1%41 / 75
BrowseCompAgentic
92.2% best2 / 63
88.7%12 / 63
Terminal-Bench 4.0Agentic
37.9% best19 / 61
31.8%23 / 61
DeepSWEAgentic
72.7% best15 / 59
67.7%27 / 59