VerdictPal
Join

Models / OpenAI

GPT-5.6 Sol

GPT-5.6 · Released 2026-07-09

OpenAI's July 2026 GPT-5.6 flagship, superseded as the current OpenAI flagship by GPT-6 Astra on 3 Sep 2026. Promo API list still $4 / $20 per 1M tokens (short context) through 21 Nov 2026. Live AA Intelligence Index v4.2 is 51 at max effort.

Previous OpenAI flagship when Astra's $10 / $50 list is too expensive. Prefer sourced Vals and AA ledger rows; live v4.2 AA scores are not comparable to the July v4.1 snapshot.

#3 of 24 on Vals Index · current snapshot
73 Vals Index · 17 benchmark rows · 5 sourcesPosition is benchmark-specific — not a cross-family or cross-source ranking.
Composite62
Context1M
Input / 1M$4.0
Output / 1M$20
Knowledge cutoffunpublished
Output speed77 t/s
TTFT (default API)91.64s
Statusflagship

Family profile

Best published score in each covered benchmark family.

79/100 avg
Knowledge
59
Reasoning
94
Coding
73
Agentic
88
Multimodal
83
Long context
78

6 tested benchmark families

Editor's note
Sol moved from limited preview to general availability on July 9, 2026. GPT-6 Astra is now the OpenAI flagship (3 Sep 2026). Live AA Intelligence Index v4.2 is 51 at max; AA lists a 1M context window. Editorial depth still needs a desk pass.
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
CritPtReasoning#1/ 1232Artificial Analysis — CritPt2026-09-01
DeepSWECoding#2/ 873DeepSWE leaderboard2026-08-04
Terminal-Bench-ScienceAgentic#2/ 922Terminal-Bench-Science 0.1 announcement2026-08-27
Vals IndexAgentic#3/ 2373Vals AI — Vals Index2026-07-16
Vending-Bench 2Agentic#3/ 10Andon Labs Vending-Bench 22026-08-04
AA-OmniscienceKnowledge#4/ 1259Artificial Analysis — AA-Omniscience2026-09-01
AA time to first tokenPerformance#4/ 186Artificial Analysis2026-09-01
MMMU-ProMultimodal#4/ 883Artificial Analysis — MMMU-Pro2026-09-01
Terminal-BenchAgentic#4/ 18488Artificial Analysis2026-09-01
GPQA DiamondReasoning#5/ 18594Artificial Analysis2026-09-01
Humanity's Last ExamReasoning#6/ 18050Artificial Analysis2026-09-01
τ³-Bench BankingAgentic#10/ 10644Artificial Analysis2026-09-01
SciCodeCoding#11/ 18056Artificial Analysis2026-09-01
AA-LCRLong context#23/ 17978Artificial Analysis2026-09-01
Artificial Analysis Intelligence IndexReasoning#36/ 18751Artificial Analysis — Intelligence Index2026-09-05
IFBenchReasoning#40/ 12773Artificial Analysis2026-09-01
AA output speedPerformance#53/ 187Artificial Analysis2026-09-01
Family context

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

KnowledgeGPT-5.6 Sol · 59
  1. 1
    GPT 5.5
    91
  2. 2
    Claude Opus 4.8
    91
  3. 3
    Claude Sonnet 4.6
    91
ReasoningGPT-5.6 Sol · 94
  1. 1
    Claude Mythos Preview
    95
  2. 2
    Gemini 3.7 Flash
    95
  3. 3
    Grok 4.6
    95
CodingGPT-5.6 Sol · 73
  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 · 78
  1. 1
    Muse Spark 1.2
    83
  2. 2
    Kimi K3
    83
  3. 3
    Gemini 3.5 Flash
    81
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.

18 recent
BenchmarkFamilyScoreSourceDays ago
Artificial Analysis Intelligence IndexReasoning51%Artificial Analysis — Intelligence Index1
CritPtReasoning32.3%Artificial Analysis — CritPt5
AA-OmniscienceKnowledge59.4%Artificial Analysis — AA-Omniscience5
MMMU-ProMultimodal83.4%Artificial Analysis — MMMU-Pro5
AA-LCRLong context77.67%Artificial Analysis5
AA output speedPerformance77 t/sArtificial Analysis5
AA time to first tokenPerformance91.64sArtificial Analysis5
Artificial Analysis Intelligence IndexReasoning60.9%Artificial Analysis5
GPQA DiamondReasoning94.1%Artificial Analysis5
Humanity's Last ExamReasoning49.5%Artificial Analysis5
IFBenchReasoning72.65%Artificial Analysis5
SciCodeCoding56.1%Artificial Analysis5
τ³-Bench BankingAgentic44.33%Artificial Analysis5
Terminal-BenchAgentic88.01%Artificial Analysis5
Terminal-Bench-ScienceAgentic22.4%Terminal-Bench-Science 0.1 announcement10
DeepSWECoding73%DeepSWE leaderboard33
Vending-Bench 2Agentic9619.37Andon Labs Vending-Bench 233
Vals IndexAgentic73.1%Vals AI — Vals Index52
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.

Knowledge59
Reasoning94
Coding73
Agentic88
Multimodal83
Long context78
Price vs. performance
0255075100$0.10$0.30$1$3$10$30input USD / 1M tokens · log scale →← intelligence

Hover any dot for name, score, and input price. Keyboard: tab through the top twelve, or use the ranking below.

AnthropicOpenAIMistralAlibabaDeepSeekGoogleMoonshotMetaMiniMaxxAIZhipuXiaomiOtherMeituanCohereNVIDIA

Full benchmark ledger

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

18 sourced rows

17 of 37 catalog benchmarks have a sourced row for GPT-5.6 Sol.

Knowledge

Reasoning

Coding

Agentic

Multimodal

Long context

Performance

Changelog
  1. Live AA Intelligence Index v4.2 is 51 at max. GPT-6 Astra is the current OpenAI flagship; Sol stays the cheaper GPT-5.6 line.
  2. Recorded general availability across ChatGPT, Codex, and the API, then refreshed Vals and Artificial Analysis benchmark evidence.
  3. 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.
  4. OpenAI previewed GPT-5.6 Sol alongside Terra and Luna.
Sources