Tasks, with the evidence
Each line names the measurement it rests on. A task that is not listed has no measurement or declared feature we can show.
Measured by a third party
Coding Index: 72.2 · Rank 8 of 86, among scored models · Top 10% · Artificial Analysis, read November 18, 2026
Code: 1,299 · Rank 17 of 122, among scored models · Top 25% · Design Arena, read November 21, 2026
Cost of this job: $0.85
Models that score at least as high for less in this scenario: 1
Measured by a third party
Agentic Index: 43.2 · Rank 6 of 64, among scored models · Artificial Analysis, read November 18, 2026
Cost of this job: $5.99
Models that score at least as high for less in this scenario: 1
Measured by a third party
Websites: 1,305 · Rank 13 of 127, among scored models · Top 10% · Design Arena, read November 21, 2026
UI components: 1,291 · Rank 21 of 118, among scored models · Top 25% · Design Arena, read November 21, 2026
Web apps (agent): 1,195 · Rank 19 of 47, among scored models · Design Arena, read November 21, 2026
Full-stack apps (agent): 1,182 · Rank 18 of 45, among scored models · Design Arena, read November 21, 2026
Mobile apps (agent): 1,174 · Rank 27 of 50, among scored models · Design Arena, read November 21, 2026
Android apps (agent): 1,207 · Rank 13 of 41, among scored models · Design Arena, read November 21, 2026
Cost of this job: $0.43
Models that score at least as high for less in this scenario: 2
Measured by a third party
Game development: 1,282 · Rank 22 of 121, among scored models · Top 25% · Design Arena, read November 21, 2026
3D scenes: 1,270 · Rank 24 of 114, among scored models · Top 25% · Design Arena, read November 21, 2026
Games (agent): 1,126 · Rank 14 of 25, among scored models · Design Arena, read November 21, 2026
Blender 3D (agent): 1,053 · Rank 15 of 25, among scored models · Design Arena, read November 21, 2026
Cost of this job: $0.43
Models that score at least as high for less in this scenario: 5
Measured by a third party
PowerPoint files (agent): 1,157 · Rank 18 of 29, among scored models · Design Arena, read November 21, 2026
HTML slides (agent): 1,034 · Rank 22 of 25, among scored models · Design Arena, read November 21, 2026
Spreadsheets (agent): 1,020 · Rank 26 of 32, among scored models · Design Arena, read November 21, 2026
Cost of this job: $0.43
Models that score at least as high for less in this scenario: 3
Measured by a third party
Data visualization: 1,326 · Rank 9 of 112, among scored models · Top 10% · Design Arena, read November 21, 2026
SVG graphics: 1,264 · Rank 18 of 84, among scored models · Top 25% · Design Arena, read November 21, 2026
ASCII art: 1,267 · Rank 10 of 68, among scored models · Top 25% · Design Arena, read November 21, 2026
Cost of this job: $0.43
Models that score at least as high for less in this scenario: 1
Declared, not measured
Context window: 1M tokens as declared. It says how much fits, not how well the model uses it.
Cost of this job: $1.55
Declared, not measured
Accepts: images, files, and video. Declared by the provider through OpenRouter; we did not test it.
Cost of this job: $0.47Text-token prices only; images, audio and video may be billed separately.
The size of each job is our assumption; the prices are real.
- Writing code: Requests: 40. Input tokens per request: 30,000. Output tokens per request: 1,500. Share of input served from cache: 70%.
- Agents that use tools: Requests: 200. Input tokens per request: 45,000. Output tokens per request: 2,500. Share of input served from cache: 75%.
- Websites and apps; Games and 3D; Slides and spreadsheets; Charts and graphics: Requests: 8. Input tokens per request: 12,000. Output tokens per request: 10,000. Share of input served from cache: 30%.
- Long documents: Requests: 3. Input tokens per request: 400,000. Output tokens per request: 4,000. Share of input served from cache: 0%.
- Images, files, audio, video in: Requests: 15. Input tokens per request: 20,000. Output tokens per request: 3,000. Share of input served from cache: 30%.
Thinking is billed as output tokens; a model that thinks longer costs more than this estimate. Ranks are among all models we list, including older ones. The share is rounded up to whole models, so a rank slightly past the exact share still counts. Models with equal scores share the better rank.