OpenAI
OpenAI is the company that launched the modern AI era with ChatGPT in November 2022. On June 26, 2026 it previewed GPT-5.6 as a three-model family (Sol, Terra, Luna) under a US Government limited-preview arrangement, then made all three tiers generally available on July 9, 2026. Pricing has moved twice since. The July 30 cut dropped Luna 80 percent to $0.20 per 1M input and $1.20 output and Terra 20 percent to $2 / $12, leaving the flagship untouched. The August 21 cut finally reached Sol: $4 per 1M input (down from $5), $20 output (down from $30), and $0.40 cached input, the first move on the flagship rate since GPT-5.5 set $5 / $30 in April 2026. OpenAI lists the Sol rate as promotional through at least November 21, 2026, and it applies to the pay-as-you-go API, Codex credits, and eligible ChatGPT Work plans but not to Pro, Plus, or Business subscriptions. In ChatGPT, Free and Go users default to Terra while paid users default to Sol. GPT-5.5 (April 2026) remains broadly available with a 1M context window at the old $5 / $30. GPT-4.5 was removed from ChatGPT on June 26, 2026, following GPT-5.2 Instant, Thinking, and Pro on June 12. On September 3, 2026 OpenAI released GPT-6 Astra, and it reverses the direction of the last two moves: $10 per 1M input and $50 output, roughly two and a half times the Sol rate, with cached input at $1, a $12.50 cache write, Batch at half price, and a Fast mode at 2x. Astra ships a 1,050,000 token context window with 128K output and an April 30, 2026 knowledge cutoff, but prompts above 272,000 input tokens bill at 2x input and cache rates and 1.5x output, so the usable window and the billed window diverge partway up. OpenAI reports 96.0 on GPQA Diamond, 97.6 on FrontierMath Tier 4, and MRCR v2 at 100 percent in the 256K to 512K band against 91.5 for Sol, with no SWE-bench Pro figure published. Rollout began as a limited partner preview, then reached paid users the next day in a restricted build that refuses certain cybersecurity prompts. On September 22, 2026 OpenAI completed the GPT-6 generation below Astra with GPT-6 Sol at $2 / $10 and GPT-6 Luna at $0.10 / $0.50 per 1M tokens, half or less of the GPT-5.6 equivalents and permanent rather than promotional, both with 1,050,000 token context and 128K output; there is no GPT-6 Terra. Independent boards place Sol at 49.3 on FrontierCode and eighth on the Vals Index, and OpenAI says Astra remains its best model across the board. OpenAI operates the largest AI developer ecosystem with broad third-party integrations, plugins, and the most recognizable consumer AI brand in the world.
Service status
Loading…
Founded
2015
Headquarters
San Francisco, CA
CEO
Sam Altman
Models
11 active
Key Products
Strengths
- ✓Largest developer ecosystem
- ✓GPT-6 Astra at a 1.05M context window with a 272K billing cliff
- ✓GPT-6 Sol and Luna at $2 / $10 and $0.10 / $0.50, permanent rates
- ✓GPT-5.6 Sol, Terra, and Luna generally available July 9, 2026
- ✓Flagship Sol cut to $4 / $20 per 1M tokens on August 21, 2026
- ✓Native audio support in GPT-4o
- ✓Strong reasoning models (o1, o3)
- ✓Consumer brand recognition
- ✓Broad enterprise partnerships
OpenAI Models
| Model | Input / 1M | Output / 1M | Context | Capabilities |
|---|---|---|---|---|
| GPT-6 Astra | 10.00 | 50.00 | 1.1M | text, vision, tool-use, code, reasoning |
| GPT-6 Sol | 2.00 | 10.00 | 1.1M | text, vision, tool-use, code, reasoning |
| GPT-6 Luna | 0.10 | 0.50 | 1.1M | text, vision, tool-use, code, reasoning |
| GPT-5.6 Sol | 4.00 | 20.00 | 1M | text, vision, tool-use, code, reasoning |
| GPT-5.6 Terra | 2.00 | 12.00 | 1M | text, vision, tool-use, code, reasoning |
| GPT-5.6 Luna | 0.20 | 1.20 | 1M | text, vision, tool-use, code, reasoning |
| GPT-5.5 | 5.00 | 30.00 | 1M | text, vision, tool-use, code, reasoning |
| GPT-4o | 2.50 | 10.00 | 128K | text, vision, tool-use, code |
| GPT-4o-mini | 0.15 | 0.60 | 128K | text, vision, tool-use, code |
| o1 | 15.00 | 60.00 | 200K | text, reasoning, code |
| o3-mini | 1.10 | 4.40 | 200K | text, reasoning, code |
Prices per 1M tokens in USD. See the full pricing guide for detailed analysis.
Benchmark Scores
| Model | SWE-bench | MMLU-Pro | HumanEval | GPQA Diamond | MATH | OSWorld 2.0 | BrowseComp | FrontierCode v1.1 | Terminal-Bench 4.0 | Humanity's Last Exam (tools) |
|---|---|---|---|---|---|---|---|---|---|---|
| GPT-6 Sol | N/A | N/A | N/A | N/A | N/A | N/A | N/A | 49.3 | N/A | N/A |
| GPT-6 Luna | N/A | N/A | N/A | N/A | N/A | N/A | N/A | 42.4 | N/A | N/A |
| GPT-5.6 Sol | 96.2 | 89.1 | N/A | 94.6 | N/A | 62.6 | 90.4 | 47.5 | 27.8 | N/A |
| GPT-5.6 Terra | 95.4 | 86.7 | N/A | 92.9 | N/A | 50.2 | 87.5 | 41.3 | 26.3 | N/A |
| GPT-5.6 Luna | 93.0 | 86.0 | N/A | 92.3 | N/A | 45.6 | 83.3 | 39.8 | 4.5 | N/A |
| GPT-5.5 | 82.6 | 94.2 | 97.1 | 78.3 | 95.8 | N/A | N/A | 43.0 | N/A | N/A |
| GPT-4o | 33.2 | 87.2 | 90.2 | 59.1 | 81.3 | N/A | N/A | N/A | N/A | N/A |
| GPT-4.5 | 38.0 | 90.1 | 93.4 | 68.7 | 88.2 | N/A | N/A | N/A | N/A | N/A |
| o1 | 48.9 | 91.8 | 94.2 | 72.5 | 94.6 | N/A | N/A | N/A | N/A | N/A |
| o3-mini | 49.3 | 86.3 | 89.7 | 60.3 | 87.1 | N/A | N/A | N/A | N/A | N/A |
| GPT-6 Astra | N/A | N/A | N/A | 96.0 | N/A | N/A | 91.5 | 53.3 | 57.1 | 57.2 |
Comparisons
Claude Opus 4.7 vs GPT-4o
Anthropic vs OpenAI
GPT-4o vs Gemini 2.5 Pro
OpenAI vs Google
GPT-4o vs Claude Sonnet 4.6
OpenAI vs Anthropic
GPT-5.5 vs Claude Opus 4.7
OpenAI vs Anthropic
GPT-5.5 vs Gemini 2.5 Pro
OpenAI vs Google
o3-mini vs Claude Sonnet 4.6
OpenAI vs Anthropic
Grok 4.6 vs GPT-5.6 Sol
xAI vs OpenAI
GPT-6 Astra vs Claude Fable 5.1
OpenAI vs Anthropic
GPT-6 Astra vs GPT-5.6 Sol
OpenAI vs OpenAI
Claude Opus 5.5 vs GPT-6 Astra
Anthropic vs OpenAI
GPT-6 Sol vs GPT-5.6 Sol
OpenAI vs OpenAI
GPT-6 Sol vs Claude Sonnet 5
OpenAI vs Anthropic
Is OpenAI Down?
Real-time status monitoring for OpenAI services, updated every 2 minutes.