AI Models

Start here AI models explained: ChatGPT, Claude, Gemini and the rest

AI Models covers the systems builders actually call: new LLM releases, open weights, image models, pricing and context windows. We track what shipped, what it costs per million tokens, and whether you can run it yourself, so you can pick a model without chasing every rumour.

What we track

Flagship and open model launches, API list prices, context length, and whether weights are available to self-host. Leaderboard rumours get labelled as such.

Models vs latest Codema coverage

Model / topic Latest Codema metric Source article
Claude Opus 5.5 $4 / $20 per million input/output tokens; 1M context Opus 5.5 for builders
GPT-6 Astra (API) $10 / $50 per million input/output tokens GPT-6 explained
Gemini 4 Argon Introductory $2 / $10 per million tokens; cached input 95% off; price doubles after Nov 30, 2026 Argon pricing
MiniMax M3.1 Flash $0.15 / $0.60 per million tokens; MiniMax reports 73.7% on DeepSWE v1.1 M3.1 Flash
Mercury 3 (diffusion) About 1,068 tokens per second (vendor framing in our piece) Diffusion LLMs
FLUX 3 Image Klein Open image model aimed at an 8GB GPU FLUX 3 Klein
Nemotron 3.5 Super, Lightning and Switchyard router explained Nemotron 3.5

Related AI articles

All Models & LLMs articles

Newest first.

More in this section