OpenAINovember 17, 2026

GPT-6.2 Sol

OpenAI GPT-6.2 Sol

OpenAI released GPT-6.2 Sol on November 17, 2026 for everyday coding and agent workflows at scale.

Text + Image inputText outputClosed / API only

At a glance

Context window
1.05M tokens
Max output
128K tokens
Text input / output · per 1M tokens
$2.00 / $10.00
Knowledge cutoff
2026-06-01

Benchmark snapshot

Best result in 6 of 8 categories

View benchmark details (20)

Benchmark details

20 results · Best published score per benchmark, grouped by category

Benchmark · settingScoreRank
Academic
GPQA Diamond

Thinking (No Tools) · Scientific Reasoning

96.04 / 200
HLE-Diamond

Max (No Tools) · Knowledge Exams

64.05 / 16
Terminal-Bench-Science 0.1

Thinking High (With Tools) · Scientific Computing

Source: Artificial Analysis (artificialanalysis.ai)
67.66 / 32
Coding
FrontierCode 2.0

Max (With Tools) · Repository Engineering

44.45 / 36
SWE-sweep

Max (With Tools) · limited comparability · Repository Engineering

11.63 / 20
Terminal-Bench 4.0

Max (With Tools) · Agentic Development

Source: Artificial Analysis (artificialanalysis.ai)
60.110 / 90
Terminal-Bench 4.1

Max (With Tools) · Agentic Development

61.56 / 48
Cybersecurity
SEC-Bench Pro

Max (With Tools) · Vulnerability Analysis

88.32 / 13
General Evaluation
AA Intelligence Index v5.0.1

Max (task-specific tools) · Capability Indices

57.36 / 62
Agent Arena Index

Max (With Tools + Internet) · Preference Arenas

Source: Arena Agent Arena Index
124110 / 36
Vals Index v2.2

Max (task-specific tools) · Capability Indices

66.19 / 36
Multimodal
Chartography

Max (No Tools) · Documents & Charts

75.114 / 29
GDP.pdf

Max (No Tools) · Documents & Charts

Source: Artificial Analysis (artificialanalysis.ai)
30.08 / 96
LVBench

Max (No Tools) · Video Understanding

87.64 / 7
Professional
AA-AnalystAgent

Max (With Tools + Internet) · Data Analysis

Source: Artificial Analysis (artificialanalysis.ai)
53.87 / 39
AutomationBench

Max (With Tools) · Private held-out (657 tasks) · Office & Business

41.325 / 64
BenchCAD 2.1 Core

Max (With Tools) · Design & Engineering

73.13 / 7
Harvey LAB-AA v1.1 (All-pass)

Max (With Tools) · Legal

Source: Artificial Analysis (artificialanalysis.ai)
11.13 / 12
Reasoning
SimpleBench

Reported best (effort unspecified) · Commonsense

Source: SimpleBench
69.930 / 106
Science & Health
HealthBench Professional

Max (No Tools) · length-adjusted · limited comparability · Clinical Workflows

65.94 / 19
10 more results in General Evaluation, Multimodal, Professional and 2 more

Rank is this model's position among every model with a published result on that leaderboard; the bar shows how far up the field that is. Orange marks first place on leaderboards with at least 5 models. Unranked leaderboards publish a composite or direction-less score.

Back to benchmark snapshot ↑

API price vs benchmark score

Pass Rate (higher is better)This modelPareto frontier34 priced models
0$020.26$6.4840.53$12.9660.79$19.44Blended API price / 1M tokens (4:1 input/output, uncached)Gemini 4 NeonGPT-6.2 Sol
GPT-6.2 Sol$3.60 / 1M41.3%Max (With Tools)
Result source
Methodology & notes (1 unpriced omitted)

Uses listed uncached text API rates at a 4:1 input/output mix against one published protocol and setting (zapier-private-heldout). API rates are separate from evaluation token costs and do not measure cost per solved task. Selected point basis: Standard pricing. Up to 272,000 input tokens Official OpenAI per-million-token pricing. Full request uses the applicable input-token band.

API pricing

Listed text rates, USD per million tokens · effective 2026-11-17

Input
$2.00
Output
$10.00
Cached input
$0.10

Blended price4:1 input:output · 90% of input tokens cached · same basis as the price ranking

$2.232/ 1M tokens

Estimate running cost for GPT-6.2 Sol

Specifications

Provider
OpenAI
Model type
Reasoning model
Released
November 17, 2026
Architecture
Undisclosed
Context window
1.05M tokens
Max output
128K tokens
Knowledge cutoff
2026-06-01
Input
Text, Image
Output
Text
Weights
Closed / API only
Commercial use
Proprietary

Capabilities

ReasoningCodingFunction callingTool useStructured outputComputer useWeb searchLong context

Model analysis

OpenAI released GPT-6.2 Sol on November 17, 2026 for everyday coding and agent workflows at scale. OpenAI launch announcement

Task difficulty and cost shape model selection

OpenAI reports that at max effort Sol was within half a point of Astra on FrontierCode 2.0 at roughly one fifth of the cost. Evaluation source

In the independent Artificial Analysis Intelligence Index v5.0.1, max effort scored about 57, at 7.9 minutes and USD 1.01 per task. For repeatable workloads, Sol offers lower rates. Astra is an option where better deliverables or fewer revisions justify higher cost. Independent model results

Coding and engineering benchmarks measure different outcomes

On Cognition’s FrontierCode 2.0, both models ran at max effort in Codex. Astra scored about 44.2% and Sol about 43.8%. On BenchCAD 2.1 Core with a customized Codex harness, Astra scored 82.5 at USD 5.43 per case and Sol scored 73.1 at USD 1.60 per case. FrontierCode · BenchCAD 2.1

In the independent Artificial Analysis Coding Agent Index with Codex at max effort, Astra is level with GPT-6 Astra and Sol trails GPT-6.1 Sol by three points, driven mainly by repository question answering. On the Vals Science Index, Astra scored 68.44% and Sol 57.76%, at USD 4.76 and USD 1.12 per test, respectively. Artificial Analysis release tests · Vals Science Index

The API supports a 1.05 million token context

gpt-6.2-sol accepts text and images and returns text. Maximum input and output are 922,000 and 128,000 tokens. The knowledge cutoff is 2026-06-01. Responses, Chat Completions, and Batch are supported, with function calling, structured outputs, file and web search, and computer use. The Responses API offers multi-agent and discounted background completion features in beta. OpenAI has not disclosed parameter counts or a language inventory. Official model documentation

Long requests and faster tiers change the bill

Standard short-context rates per million tokens are USD 2 input, 0.1 cached input, and 10 output. Above 272,000 input tokens, input and cache rates double and output costs 1.5 times the short-context rate for the full request. Batch is half the Standard price and Fast is twice the Standard price. Ultrafast costs three times Standard through February 28, 2027 and six times Standard from March 1, Pacific time. ChatGPT Work and Codex serve Plus, Pro, Enterprise, and Edu users, with Ultrafast access depending on plan and workspace settings. Official API pricing

Pro 500 supports Ultrafast on both models. Pro 200 supports it on Sol at launch. Enterprise and Edu access depends on workspace settings. ChatGPT availability details

Tool calling uses the Responses API. Eligible background requests can choose a two-hour completion window at 30% below Standard rates or an eight-hour window at 40% below Standard. OpenAI API changelog

Amazon Bedrock and Microsoft Foundry also announced general availability for both base models on launch day. Region and deployment options vary by platform. AWS announcement · Microsoft Foundry model documentation

OpenAI tightened safeguards for higher-risk requests

The system card describes deployment safeguards and remaining limitations in how the models report blockers. GPT-6.2 system card

Last reviewed 2026-11-18

GPT-6.2 Sol vs competitors

Comparable results only · record is GPT-6.2 Sol's wins–losses

GPT-6.2 Sol benchmark scores against competitors
BenchmarkGPT-6.2 SolThis modelvs GPT-6 AstraRecord 10–7vs Claude Opus 5.5Record 8–9vs Claude Fable 5.1Record 12–4vs Claude Fable 5.2Record 3–13
SimpleBench69.983.688.486.694.9
GDP.pdf30.032.228.828.026.8
AA Intelligence Index v5.0.157.356.959.15064.6
Terminal-Bench 4.161.560.365.857.665.8
Agent Arena Index12411282133612721448
FrontierCode 2.044.444.045.532.449.1
Terminal-Bench-Science 0.167.668.163.340.574.6
Chartography75.191.089.086.291.6
HealthBench Professional65.964.765.662.168.1
HLE-Diamond64.065.860.955.180.2

Lineage

Predecessors and successors · oldest to newest

  1. GPT-5.4Mar 5, 2026
  2. GPT-5.5Apr 23, 2026
  3. GPT-5.6 SolJun 26, 2026
  4. GPT-6 SolSep 22, 2026
  5. GPT-6.1 SolSep 29, 2026
  6. GPT-6.2 SolThis model · Nov 17, 2026vs GPT-6.1 Sol

Generational jump vs GPT-6.1 Sol

Showing the 2 largest gains and 2 smallest gains or drops out of 10 like-for-like benchmarks.

Full head-to-head
GPT-6.2 SolGPT-6.1 SolOther ranked models

Max (With Tools) · 12 ranked

6.9%→11.1%
0102015.0GPT-6.2 AstraMuse Wonder Fu…11.1GPT-6.2 SolMuse WonderGrok 4.7Muse Spark 1.3GPT-6 AstraClaude Fable 5…6.9GPT-6.1 SolClaude Fable 5…Kimi K3Claude Opus 5.5+4.2% · #9→#3
Hover or focus any bar to inspect rank & scoreRank #9 → #3

Max (With Tools) · 7 ranked

52.6→73.1
047.59583.5Claude Fable 5…GPT-6.2 Astra73.1GPT-6.2 SolClaude Opus 5.5GPT-6 Astra52.6GPT-6.1 SolClaude Fable 5…+20.5 · #6→#3
Hover or focus any bar to inspect rank & scoreRank #6 → #3
Agent Arena IndexSmallest gain

Max (With Tools + Internet) · 36 ranked

1241→1241
1.1K1.3K1.5K1448Claude Fable 5…Claude Opus 5.5Gemini 4 NeonGemini 4 CarbonClaude Sonnet…GPT-6.2 AstraGPT-6 AstraClaude Fable 5…Gemini 4 Hydro…1241GPT-6.2 Sol1241GPT-6.1 SolGPT-6.1 Terra+0.05 · #11→#10
Hover or focus any bar to inspect rank & scoreRank #11 → #10

Reported best (effort unspecified) · 83 ranked

82.9→69.9
05010094.9Claude Fable 5…Gemini 4 NeonGemini 4 CarbonGemini 4 ArgonGemini 4 Hydro…Claude Opus 5.5GPT-6 Astra82.9GPT-6.1 SolGemini 3.8 Fla…Grok 4.569.9GPT-6.2 SolGPT-5.5−13.0 · #10→#29
Hover or focus any bar to inspect rank & scoreRank #10 → #29
4 of 10 shared benchmarks shown

Price & benchmark trends

Across curated predecessors and successors, ordered by release date. Points use listed rates and published results, not a history of price changes within one model. Related branches may appear; adjacent points do not imply direct succession. Undated models appear last.

API price across generations

Text inputText outputUSD / 1M tokens
$0$16.2$32.4GPT-5.4GPT-5.5GPT-5.6 SolGPT-6 SolGPT-6.1 SolGPT-6.2 Sol

Input and output share one linear USD axis · uncached text rates · lower is cheaper. Retired models may retain their last recorded price. Rate tiers and context limits can differ; inspect each model's pricing conditions.

Benchmark performance across generations

0612GPT-5.4GPT-5.5GPT-5.6 SolGPT-6 SolGPT-6.1 SolGPT-6.2 Sol

All-criteria task pass rate (%) · Higher is better. Same recorded metric, protocol and setting; review exclusions are respected. Score gaps do not establish statistical significance.

Inspect Harvey LAB-AA v1.1 (All-pass) →

Orange marks this model. Hover, tap or focus the chart to see exact values. Missing values leave gaps. Horizontal spacing shows release order, not elapsed time.

Listed prices and selected benchmark scores across model generations
Model · releasedInput / 1MOutput / 1MPrice effectiveHarvey LAB-AA v1.1 (All-pass)
1GPT-5.4Mar 5, 2026$2.50$15.002026-08-30—
2GPT-5.5Apr 23, 2026$5.00$30.002026-04-23—
3GPT-5.6 SolJun 26, 2026$4.00$20.002026-08-30—
4GPT-6 SolSep 22, 2026$2.00$10.002026-09-22—
5GPT-6.1 SolSep 29, 2026$2.00$10.002026-09-296.9 ↗
6GPT-6.2 SolNov 17, 2026 · This model$2.00$10.002026-11-1711.1 ↗

Community discussion

Share the API, setup, and tradeoffs you actually tested. Rate the model once; comment as often as you like.

Used this model? Sign in with Google to rate it and share what worked, what did not, and which API or setup you tested.

No community comments yet.