Skip to content
VECTOR WIREAI INTELLIGENCE
⌘K
UTC
Models838Deals30d267Policy30d80, 15 pendingSources1,184

GPT-6 Sol

OpenAIReleased Sep 22, 2026Current

Score basis

GPT-6 Sol is strong in math, long context, and factuality; capable in reasoning and multimodal tasks; and behind the leaders in agentic tasks, coding, and instruction following. Too few results yet to rate safety or multilingual tasks.

Capability profile

Bars show the model's median result as a share of the leading model's, per capability. Select a row to see its results.

Strong3 capabilities
  1. Math−6.2%5 of 5
  2. Long Context−7.6%1 of 3
  3. Factuality−8.0%3 of 4
Capable2 capabilities
  1. Reasoning−10.5%5 of 6
  2. Multimodal−22.7%1 of 6
Limited3 capabilities
  1. Agentic−28.5%2 of 7
  2. Coding−28.7%4 of 10
  3. Instruction Following−40.5%1 of 3
Not rated2 capabilities

Too few results yet to rate Safety or Multilingual.

Price

$2.00input$10.00outputper million tokens

From Artificial Analysis · 2 providers tracked · All prices

Costs more than 82% of 331 priced models · 3:1 input-to-output blend, log scale

Evidence

117results on42benchmarks

  • 41 independently verified
  • 66 aggregator
  • 7 vendor-reported
  • 3 cross-referenced

From 16 sources · latest Nov 13, 2026 · How verification works

API features

Tool callingStructured outputsReasoning

As listed by OpenRouter

Research

18 papers reference GPT-6 Sol

GPT-6 Sol benchmark results

117 results on 42 benchmarks, grouped by capability. Each bar is the result as a share of the capability leader's; every result links to its source.

Math

StrongFull math ranking

6.2% behind the leader5 of 5 ranked benchmarks measured

Long Context

StrongFull long context ranking

7.6% behind the leader1 of 3 ranked benchmarks measured

Show 4 more long context resultsHide 4 long context results

Factuality

StrongFull factuality ranking

8.0% behind the leader3 of 4 ranked benchmarks measured

Show 11 more factuality resultsHide 11 factuality results

Reasoning

CapableFull reasoning ranking

10.5% behind the leader5 of 6 ranked benchmarks measured

Show 27 more reasoning resultsHide 27 reasoning results

Multimodal

CapableFull multimodal ranking

22.7% behind the leader1 of 6 ranked benchmarks measured

Show 6 more multimodal resultsHide 6 multimodal results

Agentic

LimitedFull agentic ranking

28.5% behind the leader2 of 7 ranked benchmarks measured

Show 11 more agentic resultsHide 11 agentic results

Coding

LimitedFull coding ranking

28.7% behind the leader4 of 10 ranked benchmarks measured

Show 5 more coding resultsHide 5 coding results

Instruction Following

LimitedFull instruction following ranking

40.5% behind the leader1 of 3 ranked benchmarks measured

More results

Benchmarks outside the capability baskets. They are not ranked against a leader.

Show 27 more resultsHide 27 results

GPT-6 Sol: common questions

Who makes GPT-6 Sol?

GPT-6 Sol is made by OpenAI.

When was GPT-6 Sol released?

GPT-6 Sol was released on Sep 22, 2026, according to Artificial Analysis.

What is GPT-6 Sol good at?

GPT-6 Sol is strong in math, long context, and factuality; capable in reasoning and multimodal tasks; and behind the leaders in agentic tasks, coding, and instruction following. Too few results yet to rate safety or multilingual tasks.

How much does GPT-6 Sol cost?

GPT-6 Sol costs $2.00 per million input tokens and $10.00 per million output tokens, according to Artificial Analysis. We track its price at 2 providers. At a mix of three input tokens to one output token, it costs more than 82% of the 331 priced models we track.

How many benchmarks has GPT-6 Sol been tested on?

We track 117 results for GPT-6 Sol on 42 benchmarks from 16 sources, 41 of them independently verified. The latest was recorded on Nov 13, 2026.

Which API features does GPT-6 Sol support?

OpenRouter lists tool calling, structured outputs, and reasoning for GPT-6 Sol.

About this record

Where GPT-6 Sol's numbers come from, and every name it appears under.

Tracked since
Sep 16, 2026
Newest source mention
Sep 29, 2026

Where the results come from

Verification: 117 scores · 41 independently verified · 66 aggregator-attributed · 3 vendor cross-reference · 7 vendor-reported. How these tiers are assigned

From 16 sources on 11 sites. Artificial Analysis supplies 66 of them; the 41 independently verified results come from 6 sites. Bars are coloured by trust tier.

  • artificialanalysis.ai66
  • arcprize.org24
  • livebench.ai7
  • deploymentsafety.openai.com4
  • raw.githubusercontent.com4
  • api.llm-stats.com3
  • epoch.ai3
  • assets.claude.com2
  • datasets-server.huggingface.co2
  • anthropic.com1
  • simple-bench.com1

Also known as

How our sources name GPT-6 Sol at each reasoning setting.

SettingShort formAPI id
lowgpt-6 sol (low)gpt-6-sol-low
mediumgpt-6 sol (medium)gpt-6-sol-medium
highgpt-6 sol (high)gpt-6-sol-high
xhighgpt-6 sol (xhigh)gpt-6-sol-xhigh
maxgpt-6 sol (max)gpt-6-sol-max
Also listed asgpt-6-sol-non-reasoninggpt-6-sol:batchgpt-6 sol (non-reasoning)gpt sol 6gpt-6 sol (none)