Usage rankings

Models by task

Seven-day request and token shares from a sampled classifier across Code, Data, Agent, and General. These are not absolute volumes or global demand.

Classified sample

Task

Classified-sample requests35.5%

Classified-sample tokens11.4%

Within category58.5%

Category tokens35.5%

Classification · Top models

1
TypeSafe: Jev 1.13typesafe/jev-1.13-20260917
20.2% req9.3% tok
2
TypeSafe: Jev 1.14typesafe/jev-1.14-20261014
12.0% req9.6% tok
3
openai/gpt-6-luna-20260922openai/gpt-6-luna-20260922
7.0% req5.8% tok
4
Perplexity: Decider V1.2perplexity/pplx-decider-v1.2-20261016
6.0% req5.6% tok
5
deepseek/deepseek-v4.1-flash-20261017deepseek/deepseek-v4.1-flash-20261017
4.7% req12.9% tok
6
deepseek/deepseek-v4.1-flash-20260910deepseek/deepseek-v4.1-flash-20260910
4.0% req4.3% tok
7
z-ai/glm-5.3-flash-20260826z-ai/glm-5.3-flash-20260826
2.8% req5.3% tok
8
Crouton Alphastealth/crouton-alpha
2.7% req5.6% tok
9
anthropic/claude-haiku-5.5-20261007anthropic/claude-haiku-5.5-20261007
2.5% req1.2% tok
10
deepseek/deepseek-v4-flash-20260731deepseek/deepseek-v4-flash-20260731
2.4% req1.8% tok

Task details

Source: OpenRouter (openrouter.ai/rankings), as of 2026-11-11.

Scope and sources

LastAI uses documented stable APIs only. The 10K context bucket, tool-using requests, and image-input activity are shown as observed token floors. Korean or Python-specific tags, tool-call counts, and image counts lack stable APIs, so proxies are separated and never used in a composite score.

Source: OpenRouter (openrouter.ai/rankings), as of 2026-11-12T05:20:08.915Z.

Methodology · external sources