ChatGPT 5.6 Luna
OpenAI
The latest generation of OpenAI's flagship model: stronger reasoning and long-term memory.
ProprietaryTesting in progress…
Test bench
Tests & benchmarks
Every test follows the same protocol so results stay comparable over time: a fixed set of tasks, the exact model version, the date, and per-task scoring. The first full test reports are being finalized.
Our methodology
- A fixed task set per category (writing, reasoning, code, images…), identical for every tool.
- The exact model version and test date recorded on every result — tools evolve fast.
- Each task scored /10 by two reviewers; disagreements are re-run and discussed.
- Outdated tests are flagged with a banner and re-run at least every 90 days.
Currently on the bench
Gemini 3 Pro Preview
The preview of Google's next pro model: native multimodality and an extended context window.
ProprietaryTesting in progress…
Qwen 3.8
Alibaba Cloud
Alibaba's open-weight model, a community-benchmark reference, deployable from mobile to server.
Open weightTesting in progress…
Mistral Large 3
Mistral AI
Mistral AI's frontier model: top-tier performance with European hosting.
ProprietaryTesting in progress…
Fable 5
Anthropic
The first model of Anthropic's Claude 5 family, in the new Mythos class above Opus in capability.
ProprietaryTesting in progress…
Veo 3
Google's video generation model: realistic clips with synchronized sound from a text prompt.
ProprietaryTesting in progress…
Nano Banana
Google's viral image editing model: consistent characters and precise edits in natural language.
ProprietaryTesting in progress…
GPT Image 2
OpenAI
OpenAI's next image model: sharper text rendering and finer instruction following.
ProprietaryTesting in progress…
Meanwhile, our verified tool sheets carry the editorial scores: browse the directory.