Compare Models

GPT-6.1 Sol

Release Date: Sep 29, 2026

Developer OpenAI  🇺🇸
Context Window 1M
Max Output Tokens 128k
Token Costs (in/out) $2.00/10.00
Weights Private
Input Modalities

Accuracy

61.15 % ± 1.01

Cost / Test (Vals Index)

$ 3.237

Latency

43 min 10 s

Vals Index
BenchmarksAccuracyRankings
Vals Index

61.15%

±1.01
7/30
Code Migration

65.12%

±4.36
4/70
CyberBench v1.1

39.29%

±2.74
40/42
EMB

70.77%

±2.71
11/68
Finance Agent (v2)

52.03%

±0.31
30/72
Legal Research Bench

38.46%

±3.38
29/71
MedCode

48.84%

±2.12
25/103
MedScribe

86.45%

±1.90
20/105
MysteryMechanism

46.40%

±3.35
5/20
ProofBench v1.1

99.00%

±1.00
4/43
SAGE

46.54%

±3.46
34/89
Public Benefits Bench v1.1

59.34%

±1.28
30/44
SRE Bench

50.76%

±3.10
2/15
Tax Agent Bench

62.31%

±3.20
26/63
Vibe Code Bench v1.1

88.93%

±1.89
6/105
BioMysteryBench

79.63%

±0.74
2/20
Harvey's Legal Agent Benchmark

5.42%

±1.75
29/72
IOI

96.89%

±3.11
2/37
Terminal-Bench 4.0

55.05%

±1.82
5/31
Proprietary BenchmarksAcademic BenchmarksIndustry Partners
Vals
Default Provider : OpenAI
Temperature: Default
Top P: Default
Top K: Default
Max Output Tokens: 128,000
Reasoning Effort: max