Open beta Zabio is in open beta: anyone can sign up and use it in production. Models, prices and features can still change while we learn from the first customers. Tell us what breaks at hello@zabio.ai.
All news
openaiAnnouncement

OpenAI fixes image encoding bug in GPT-6 Sol and GPT-6 Luna

The update improves results on visual tasks in the API and Codex, including computer use.

On Zabio today

Start free, no card needed

What changed

OpenAI fixed a bug in image encoding that degraded image understanding in GPT-6 Sol and GPT-6 Luna. The update improves results on visual tasks in the API and Codex, including computer use. OpenAI recommends rerunning evaluations and retrying workflows affected by the issue when use cases involve image inputs.

What it means for Zabio customers

If you send images to GPT-6 Sol or GPT-6 Luna, you get better results without changing models or code.

Source

Fixed a bug in image encoding that degraded image understanding in GPT-6 Sol and GPT-6 Luna. This update improves results on visual tasks in the API and Codex, including computer use.

Original announcement: OpenAI API changelog

How it compares

The models in this story next to models of the same class at a similar price, all available with one Zabio key.

ModelInput, $/MOutput, $/MCached, $/MContext window1,000 callsAnalysis and reasoning, hard
gpt-6-solOpenAIIn this story2.0010.000.20400,000$9.0094.3
gpt-6-lunaOpenAIIn this story0.100.500.01400,000$0.4573.6
gemini-4-argonGoogle2.0010.00no data200,000$9.0092.8
kimi-k3.1-proMoonshot AI2.0010.000.201,048,576$9.0086.6
qwen3.7-maxQwen2.507.500.501,000,000$8.7590.9

Prices are the vendor list prices in USD per million tokens. "1,000 calls" is 1,000 requests of 2,000 input tokens and 500 output tokens each, without caching. Zabio bills the list price plus its service fee: see pricing.

Scores: percent correct for that kind of job, aggregated from public results by Epoch AI, LMArena, and LiveBench, read on 16 November 2026. Epoch AI: CC BY 4.0, LMArena: CC BY 4.0, LiveBench: Leaderboard results used with attribution; the 2.0 release states no terms of its own (read 2026-11-06); the first version's Mathematics, Language, IF results: Code Apache 2.0 and MIT; leaderboard results used with attribution

Source: OpenAI API changelog (2026-09-25)