By the toolist.ai editors. Researched from public sources, not hands-on tested by toolist.ai. Updated October 13, 2026; prices checked October 13, 2026.

GPT-6.1 Sol vs GPT-6 Sol

Verdict

Switch to GPT-6.1 Sol. It costs the same $2 and $10 per million tokens, halves the cached-input price to $0.10, and scores higher on seven of the eight Artificial Analysis benchmarks both have. Stay on GPT-6 Sol only if you rely on the "none" effort level or need its fast output at max effort (about 123 tokens a second, per Artificial Analysis).

Verdict based on published prices, limits and benchmarks, not on our own tests.

Choose GPT-6.1 Sol if

  • You want the stronger model at the same price (51.8 against 47.6 on the Intelligence Index)
  • Your prompts repeat: cached input costs $0.10 per million tokens against $0.20
  • You run coding agents (56.1% against 43.9% on Terminal-Bench 4.0)

Choose GPT-6 Sol if

  • Your calls use the "none" effort level, for example function calling through Chat Completions; GPT-6.1 Sol dropped it
  • You need fast output at max effort, where Artificial Analysis measures GPT-6 Sol at about 123 tokens a second
  • Your work leans on scientific coding, the one shared benchmark it leads (57.6% against 54.2% on SciCode)

Price

Input costs the same on both. Output costs the same on both.

Per million tokensGPT-6.1 SolGPT-6 Sol
Input$2 source$2 source
Cached input$0.10 source$0.20 source
Output$10 source$10 source

GPT-6.1 Sol: Prompts over 272,000 input tokens are billed at twice the input rate and 1.5 times the output rate for the whole request. Batch and Flex processing cost half; Fast mode costs double and Ultrafast six times.

GPT-6 Sol: Prompts over 272,000 input tokens are billed at twice the input rate and 1.5 times the output rate for the whole request. Batch and Flex cost half; Fast mode costs double.

Worked example: what a month costs

We priced three monthly workloads at the standard rates. The provider does not publish these figures.

WorkloadGPT-6.1 SolGPT-6 Sol
Support chatbot
100,000 replies, each 3,000 input and 400 output tokens, no caching (300M in, 40M out)		 $1,000$1,000
Document Q&A
5,000 questions over the same 60,000-token document set (90% cache hits), 800 output tokens each (300M in, 4M out)		 $127$154
Coding agent
200 tasks, each 1.5 million input tokens (90% cache hits) and 50,000 output tokens (300M in, 10M out)		 $187$214

Price your own workload

ModelCost a month
GPT-6.1 Sol $150
GPT-6 Sol $150

Standard API prices, no batch or priority discounts. Cached input is billed at the cache-read price; cache-write charges are not included. Reasoning models bill their thinking as output tokens.

Limits and release dates

GPT-6.1 SolGPT-6 Sol
ReleasedSeptember 29, 2026 sourceSeptember 22, 2026 source
Context window1,050,000 tokens source1,050,000 tokens source
Maximum output128,000 tokens source128,000 tokens source
Knowledge cutoffApril 30, 2026 sourceApril 20, 2026 source
Input and outputText and images in; text out sourceText and images in; text out source
API model namegpt-6.1-solgpt-6-sol

Benchmarks

Scores as published by the provider or measured by Artificial Analysis; higher is better. A dash means no score was published.

BenchmarkGPT-6.1 SolGPT-6 Sol
Artificial Analysis Intelligence Index
max effort		 51.8 source47.6 source
Intelligence Index at everyday effort
medium effort, the API default		 47.8 source39.8 source
Coding Agent Index
Codex harness, max effort		 60.1 source56.7 source
Terminal-Bench 4.0
max effort		 56.1% source43.9% source
Humanity's Last Exam
max effort		 52.9% source47.9% source
GDPval-AA v2.1 (Elo)
max effort		 1592 source1508 source
SciCode
max effort		 54.2% source57.6% source
Terminal-Bench-Science 0.1
max effort		 58.1% source30.0% source
Output speed 55.2 tokens/s (medium effort, OpenAI API) source122.7 tokens/s (max effort, OpenAI API; no medium-effort figure is published) source

Benchmark and speed data from Artificial Analysis where linked.

Read more

Sources