OpenAI o3 is a family of reasoning-focused large language models developed by OpenAI and the second generation of the company's o-series reasoning models, best known for scoring 87.5% on the ARC-AGI abstract-reasoning benchmark in December 2024 (versus about…
This is a list of open-weight models: artificial intelligence models whose trained parameters (weights) the developer has published for download, so that the model can be run, studied and in most cases fine-tuned outside the developer's own service.
The Gemini app is Google's consumer AI assistant, a chat application available on the web at gemini.google.com and as native Android, iOS, macOS and Windows apps, powered by the Gemini family of large language models from Google DeepMind.
Gemini 4 Neon is a multimodal reasoning model from Google DeepMind, announced on November 18, 2026 for software engineering, analytical work, and scientific tasks. Google adapted it from Gemini 4 Carbon.
Antigravity is an agentic software development platform from Google, launched in public preview on 18 November 2025 alongside the release of gemini 3 pro.
This page compares open-source coding agents: AI coding tools whose own source code is published under a licence approved by the Open Source Initiative (OSI), so that anyone can read, modify and redistribute the agent itself.
Tokenization is the process of breaking text into smaller units called tokens, which serve as the fundamental input to natural language processing (NLP) systems and large language models (LLMs).
AI Compute Tracker reviewed standard (non-batch) provider list prices in USD per 1M tokens, at the short-context or peak-hour tier. Cached input is shown only when recorded in that row; it is not inferred from a discount. Long-context, cache writes, batch, tools, media and other charges are not included.
Higher is better, not an IQ or percentage. Catalogue metadata does not disclose the index version, reasoning configuration, or evaluation date; checked time is retrieval time. Treat comparisons as indicative.
Benchmark version: not supplied. Test configuration: not supplied.
AI Compute Tracker reviewed standard (non-batch) provider list prices in USD per 1M tokens, at the short-context or peak-hour tier. Cached input is shown only when recorded in that row; it is not inferred from a discount. Long-context, cache writes, batch, tools, media and other charges are not included. Tracker row notes: ≤272K prompt tokens; above 272K $4/$15
Higher is better, not an IQ or percentage. Catalogue metadata does not disclose the index version, reasoning configuration, or evaluation date; checked time is retrieval time. Treat comparisons as indicative.
Benchmark version: not supplied. Test configuration: not supplied.
AI Compute Tracker reviewed standard (non-batch) provider list prices in USD per 1M tokens, at the short-context or peak-hour tier. Cached input is shown only when recorded in that row; it is not inferred from a discount. Long-context, cache writes, batch, tools, media and other charges are not included. Tracker row notes: Current price through Nov 30, 2026; $4/$20 from Dec 1, 2026; batch $2/$10 from Dec 1, 2026
Higher is better, not an IQ or percentage. Catalogue metadata does not disclose the index version, reasoning configuration, or evaluation date; checked time is retrieval time. Treat comparisons as indicative.
Benchmark version: not supplied. Test configuration: not supplied.
AI Compute Tracker reviewed standard (non-batch) provider list prices in USD per 1M tokens, at the short-context or peak-hour tier. Cached input is shown only when recorded in that row; it is not inferred from a discount. Long-context, cache writes, batch, tools, media and other charges are not included. Tracker row notes: Peak-hour cache-miss input; off-peak $0.15/$0.60; deepseek-flash alias
Weights: whether the model's weights can be downloaded and run on your own hardware, and under which license. A model with downloadable weights can be served by other providers or self-hosted, usually at a different price than the API rate shown. Browse every open-weight model.
Example cost: 1,000 requests, each with 2,000 input and 500 output tokens, at the listed standard text rates. No caching, batch discount, long-context surcharge, tools, media or tax; not a capability or availability check. Adjust this workload in the cost planner (choose a model there).