Post-Cutoff
Search For AI agents

OpenAIGPT-6 familyCurrentReasoning modelReleased 29 Sep 2026

GPT-6.1 Sol

OpenAI’s default mid-tier model for coding agents, computer use and document work as of Sept 29, 2026.

Model idgpt-6.1-sol

Price$2 in, $10 out per 1M tokens

Checked11 October 2026

Knows up to April 2026

1,863events after its cutoff

83 historic, 426 major. 197 days of news.

It launched on 29 September 2026, already missing 570 of them.

Five things it has never heard of

How to call it

5 ways to use GPT-6.1 Sol, with the model id each provider expects.

Providers, model ids, endpoints and docs for GPT-6.1 Sol
ProviderModel idEndpointDocs
OpenAI API gpt-6.1-sol https://api.openai.com/v1/responses developers.openai.com
OpenRouter openai/gpt-6.1-sol openrouter.ai
GitHub Copilot github.blog
AWS Bedrock (GA late Sept 2026; Ultrafast mode added Oct 9; reasoning summaries via Responses API) aws.amazon.com
Web app chatgpt.com

Pricing

Prices for GPT-6.1 Sol
Input$2 per 1M tokens
Cached input$0.10 per 1M tokens
Output$10 per 1M tokens
Cache write$2.50 per 1M tokens

As published: “per 1M tokens (USD), standard tier”. Note: prompts >272K input tokens billed at 2x input/cache and 1.5x output; Fast mode 2x; Batch/Flex 50% off; Ultrafast (since Oct 8 2026) 6x standard, i.e. $12 input / $0.60 cached / $15 cache write / $60 output per 1M short-context ($24/$1.20/$30/$90 long-context).

Source: developers.openai.com, checked 11 October 2026.

Specs

Input
Text and images
Output
Text
Context window
1,050,000 tokens
Max output
128,000 tokens
Open weights
No
Licence
Proprietary
Training cutoff
April 2026
Released
29 September 2026
Status
Current
Type
Reasoning model

What stands out

What its maker marketed at launch, and what people found later, each with its source.

  • Near-Astra quality at one-fifth the price

    OpenAI says it nearly matches GPT-6 Astra on agentic coding (DeepSWE v1.1), computer use (OSWorld 2.0, within 2.1 pts at ~1/7 cost/task) and professional work at $2/$10 vs Astra’s $10/$50.

    Source: openai.com

  • Near-Astra ARC-AGI-3 score at lower cost

    Found after launch

    ARC Prize verified (Sept 30, 2026): ARC-AGI-3 96.4% at $4.4K with the provider-adapter harness (52.7% with the standard harness) vs GPT-6 Astra’s 99.9%, at 77% lower cost; ARC-AGI-2 94.2% at $0.25/task.

    Source: x.com

  • 95% cached-input discount

    Cached input costs $0.10 per 1M tokens (5% of the input rate), half of GPT-6 Sol’s cached price.

    Source: developers.openai.com

  • Improved alignment vs GPT-6 Sol

    Failed to disclose a broken search tool in 2.1% of adversarial tests (GPT-6 Sol 4.9%, Astra 1.5%); no observed attempts to bypass the automated safety reviewer.

    Source: cdn.openai.com

  • Full hosted tool suite

    Responses API tools: web/file search, image generation, code interpreter, hosted shell, apply patch, skills, computer use, MCP, tool search.

    Source: developers.openai.com

  • Multi-agent in the Responses API (beta)

    The model can spin up and coordinate parallel subagents inside one Responses API request and synthesize their results; enable with the beta header OpenAI-Beta: responses_multi_agent=v1 (also available for all GPT-5.6 models). OpenAI’s Oct 2 2026 GPT-6 model guide names GPT-6.1 Sol as the family member that supports delegation.

    Source: developers.openai.com

  • Ultrafast service tier

    Found after launch

    Since Oct 8 2026, service_tier “ultrafast” in the Responses API (all API users, separate rate limits, WebSockets recommended) at 6x standard price; also in Codex and ChatGPT Work for Pro $500 and eligible Enterprise/Edu plans. OpenAI promised up to 8x faster token generation in Codex and up to 6x in the API.

    Source: developers.openai.com

Four assistants, one scale

GPT-6.1 Sol is added as a fifth row. Each row runs from what the model knows (graphite) through the months between its cutoff and its launch (hatched) to today (pink).

  1. GPT-6 Astra 271 at launch, 1,863 today
  2. Claude Opus 5.5 372 at launch, 1,795 today
  3. Gemini 4 Hydrogen 1,863 at launch, 1,889 today
  4. Grok 4.7 387 at launch, 1,829 today
  5. GPT-6.1 Sol 570 at launch, 1,863 today
Show as a table
ModelCutoffReleasedMissing at launchMissing today
GPT-6 AstraApril 20263 September 20262711,863
Claude Opus 5.5June 202622 September 20263721,795
Gemini 4 HydrogenMarch 202612 November 20261,8631,889
Grok 4.7May 202621 September 20263871,829
GPT-6.1 SolApril 202629 September 20265701,863

Notes

Successor to GPT-6 Sol, released one week later at DevDay 2026. reasoning.effort low/medium(default)/high/xhigh/max (no none/minimal). Max input 922K tokens. Chat Completions supported without tool calling. US/EU data residency (Fast mode unavailable with EU residency). In ChatGPT Work and Codex for Plus and above; not yet in Chat as of launch. Ultrafast version promised ‘in the coming days’ at launch; it shipped on Oct 8 2026 (API changelog, Codex changelog), see OpenAI ships Ultrafast mode for GPT-6.1 Sol in the API, Codex and ChatGPT Work. Since Oct 7 2026, an October update of GPT-6 Sol (not 6.1) powers the ChatGPT Chat tab for paid tiers, see OpenAI brings GPT-6 to all ChatGPT users with ‘Intelligent UI’ answers built from interactive components. OpenRouter also lists openai/gpt-6.1-sol-pro.

curl https://api.openai.com/v1/responses \
 -H "Authorization: Bearer $OPENAI_API_KEY" -H "Content-Type: application/json" \
 -d '{"model":"gpt-6.1-sol","reasoning":{"effort":"medium"},"input":"Hello"}'

Rate limits (standard): Tier 1 500 RPM / 500K TPM … Tier 5 15,000 RPM / 40M TPM.

Sources: