Skip to content
PricingSaaS
Sign in

AI Brief

AI briefs

One researched brief per week on AI leaders and AI-priced companies. Newest first.

  1. Week of Nov 2·Nov 2 – Nov 8·6 leaders · 29 sources

    Inference reprices by the GPU while apps fold the agent into the tier

    Cohere swapped per-model Model Vault rates for a per-GPU card, Together put Qwen3.7-Max back to $2.50/$7.50 per 1M tokens, and LogicMonitor made Edwin AI standard in all three Catchpoint-successor tiers. Bundling is now 53% of AI monetization moves.

    CohereTogether AICatchpointAircallAppcuesOpenAI
  2. Week of Oct 26·Oct 26 – Nov 1·5 leaders · 33 sources

    Cheaper tokens, vaguer allowances: AI pricing moves in two directions

    Cartesia halved its voice agent rate and Baseten published prices it had hidden, while Canva pulled the number out of its free AI line and OpenAI bundled up to $500 of API usage into a Pro seat. Credits appeared in 423 diffs this period, up from 271.

    CartesiaBasetenOpenAIGitHubCanva
  3. Week of Oct 19·Oct 19 – Oct 25·5 leaders · 34 sources

    The sticker holds, the meter moves: AI pricing slid off the plan card this week

    Microsoft held every Copilot list price and began billing advanced AI as used. Cohere cut its hosted Rerank line. Composio metered a security add-on per token. Dropzone dropped a cap. In a record 203 AI-relevant changes, usage and credits drove 30% of the mix.

    MicrosoftCohereComposioDropzone AIHighlight AI
  4. Week of Sep 21·Sep 21 – Sep 27·6 leaders · 56 sources

    Cheaper flagships, metered everything else: this week in AI pricing

    Anthropic cut Opus pricing 20% and moved Opus 5 to legacy, Fireworks and Baseten added GLM 5.3 rows while retiring older models, Fastly put a per-request price on AI routing, and Notion and Aircall reworked how AI work is billed. Our capture logged 107 AI-relevant changes.

    AnthropicFireworksBasetenFastlyNotionAircall
  5. Week of Aug 24·Aug 24 – Aug 30·5 leaders · 25 sources

    The week AI vendors stopped averaging the heavy user into one price

    OpenAI split Business into a $20 seat and a $100 seat; Vercel cut Connect's included quota to 500 requests and raised the unit rate tenfold; Cohere put Parse 5 on $4.00/hr instances. Usage, credits and tokens carry 28% of AI monetization in our index.

    OpenAIMicrosoftCohereVercelIntercom