VerdictPal
Join

Models / OpenAI

GPT-5.6 Sol

GPT-5.6 · Released 2026-07-09

OpenAI's generally available GPT-5.6 flagship for coding, science, cybersecurity, computer use, and long-horizon agentic work, priced at $5 / $30 per 1M tokens.

GPT-5.5 successor for frontier coding, research, computer-use, and multi-agent workflows, now with public benchmark rows from OpenAI, Vals, and Artificial Analysis.

#4 of 21 on Vals Index · current snapshot
73 Vals Index · 11 benchmark rows · 2 sourcesPosition is benchmark-specific — not a cross-family or cross-source ranking.
Composite66
ContextUnlisted
Input / 1M$5.0
Output / 1M$30
Knowledge cutoffunpublished
Output speed67 t/s
TTFT (default API)76.91s
Statusflagship
ReasoningCodingAgenticLong context
Reasoning
94
Coding
56
Agentic
88
Long context
74

78average across 4 tested families

Editor's note
Sol moved from limited preview to general availability on July 9, 2026. The ledger now includes independent Vals and Artificial Analysis rows; OpenAI still has not published the context limit.
Benchmark placements

Where GPT-5.6 Sol places on each public benchmark source that publishes a row. Rank counts every model with a latest row in the same test — not a universal quality score.

BenchmarkFamilyRankScoreSourceDate
Terminal-BenchAgentic#1/ 6988Artificial Analysis2026-07-21
Artificial Analysis Intelligence IndexReasoning#2/ 6759Artificial Analysis2026-07-21
Humanity's Last ExamReasoning#2/ 6547Artificial Analysis2026-07-21
GPQA DiamondReasoning#3/ 6994Artificial Analysis2026-07-21
Vals IndexAgentic#4/ 2073Vals AI — Vals Index2026-07-16
AA time to first tokenPerformance#6/ 69Artificial Analysis2026-07-21
AA-LCRLong context#8/ 6474Artificial Analysis2026-07-21
SciCodeCoding#8/ 6556Artificial Analysis2026-07-21
τ³-Bench BankingAgentic#21/ 6433Artificial Analysis2026-07-21
IFBenchReasoning#27/ 5773Artificial Analysis2026-07-21
AA output speedPerformance#42/ 69Artificial Analysis2026-07-21
Family context

The three highest-scoring carded models in each capability family. Where GPT-5.6 Sol shows up, it's highlighted.

ReasoningGPT-5.6 Sol · 94
  1. 1
    Claude Mythos Preview
    95
  2. 2
    GPT-5.6 Sol
    94
  3. 3
    Kimi K3
    94
CodingGPT-5.6 Sol · 56
  1. 1
    GPT 5.5
    96
  2. 2
    Claude Opus 4.8
    96
  3. 3
    Claude Opus 4.6
    96
AgenticGPT-5.6 Sol · 88
  1. 1
    GPT 5.5
    100
  2. 2
    Claude Opus 4.8
    100
  3. 3
    Claude Fable 5
    100
Long contextGPT-5.6 Sol · 74
  1. 1
    Kimi K3
    75
  2. 2
    GPT-5.1
    75
  3. 3
    GPT-5.6 Sol
    74
Newest receipts

Benchmark rows added to the public ledger for GPT-5.6 Sol in the last 120 days. Older rows live in the full table below.

11 recent
BenchmarkFamilyScoreSourceDays ago
AA-LCRLong context73.67%Artificial Analysis2
AA output speedPerformance67 t/sArtificial Analysis2
AA time to first tokenPerformance76.91sArtificial Analysis2
Artificial Analysis Intelligence IndexReasoning58.9%Artificial Analysis2
GPQA DiamondReasoning94.1%Artificial Analysis2
Humanity's Last ExamReasoning47.2%Artificial Analysis2
IFBenchReasoning72.65%Artificial Analysis2
SciCodeCoding56.1%Artificial Analysis2
τ³-Bench BankingAgentic32.99%Artificial Analysis2
Terminal-BenchAgentic88.01%Artificial Analysis2
Vals IndexAgentic73.1%Vals AI — Vals Index7
Family coverage

A high composite that hides a weak family is a trap. These bars surface the families where this model hasn't been publicly tested, and where it leads.

Reasoning94
Coding56
Agentic88
Long context74
Price vs. performance
0255075100$0.10$0.30$1$3$10$30input USD / 1M tokens · log scale →← intelligence

Hover or tab any dot for name, score, and input price.

AnthropicOpenAIMistralAlibabaDeepSeekGoogleMoonshotMiniMaxxAIMetaZhipuXiaomiMeituanCohere

Full benchmark ledger

Every catalog benchmark for GPT-5.6 Sol. Scores link to the original source; gaps mean no public row exists yet.

11 sourced rows

11 of 32 catalog benchmarks have a sourced row for GPT-5.6 Sol.

Reasoning

Coding

Agentic

Long context

Performance

Changelog
  1. Recorded general availability across ChatGPT, Codex, and the API, then refreshed Vals and Artificial Analysis benchmark evidence.
  2. Updated public pricing: Sol is listed at $5 input / $30 output per 1M tokens, with 1.25x cache-write billing and 90% cached-read discount.
  3. OpenAI previewed GPT-5.6 Sol alongside Terra and Luna.
Sources