Avg blended cost · $/M tokens
Mistral Large 4 and MiniMax M3.1 Flash join the basket; the index eases to $1.97.
Track what's moving frontier AI costs.
Weekly index update + analysis, free. No spam, unsubscribe anytime.
TPI — Avg blended cost over time
TPI movements reflect the real cost of accessing frontier AI — rising as more capable models define the market, falling as cheaper alternatives gain traction.
Weekly pricing news
Updated October 23, 2026Mistral Large 4 replaces Large 3 at its list price
The launch discount ended on 20 October. The standing rate is well above Large 3's, but the new model scores far higher on the Artificial Analysis Intelligence Index.
MiniMax adds a Flash tier
M3.1 Flash sits beside M3 at half the price and outscores it, so the cheap end of the basket gets another strong contender.
DeepSeek refreshes V4.1 Flash at the same price
A new checkpoint went live behind the existing endpoint on 17 October, lifting its score from 40 to 45 with no change to what developers pay.
Weekly pricing news
Updated October 23, 2026Mistral Large 4 replaces Large 3 at its list price
The launch discount ended on 20 October. The standing rate is well above Large 3's, but the new model scores far higher on the Artificial Analysis Intelligence Index.
MiniMax adds a Flash tier
M3.1 Flash sits beside M3 at half the price and outscores it, so the cheap end of the basket gets another strong contender.
DeepSeek refreshes V4.1 Flash at the same price
A new checkpoint went live behind the existing endpoint on 17 October, lifting its score from 40 to 45 with no change to what developers pay.
From the newsletter
Weekly index analysis, free. and never miss a move.
Model Positioning Map
Price vs capability tier across all TPI index members. Compare models in detail →
Fast — speed & cost·Mid — balanced·Frontier — highest capability
Floor and ceiling
How the gap between the cheapest and most expensive index members has evolved through 2026. Compare models in detail →
Models in index = number of AI models tracked. Multiplier = ratio of the most expensive blended price to the cheapest.
Cost Per Task
Costs calculated at TPI average · Input $2.01/M · Output $9.53/M · Use the Simulator to model your specific workload.
Email summary
Summarize a typical business email
$0.0020
↳ 500 in / 100 out tokens
Blog post
Generate a 1000-word blog article
$0.0136
↳ 340 in / 1,360 out tokens
Code review
Review 200 lines of code
$0.0141
↳ 3,200 in / 800 out tokens
Chat conversation
Average multi-turn chat session
$0.0125
↳ 1,500 in / 1,000 out tokens
SQL generation
Generate SQL from natural language
$0.0098
↳ 850 in / 850 out tokens
RAG query
Answer a question using retrieved context
$0.0108
↳ 3,000 in / 500 out tokens
Token counts shown are upper-end estimates for typical production workloads. Actual usage varies by prompt structure, context size, and use case.