Previous edition: Claude model share by coding agent, April 2026
Claude model share by coding agent, October 2026
Claude model share by coding agent, October 2026
Percentage of total spend allocated to Claude models (Sonnet, Opus, Haiku) per agent.
Claude share
In October 2026 Claude models took 92% of Qwen Code spend and 89% of Claude Code spend.
Which models do coding agents spend on? In October 2026 Claude models took 92% of Qwen Code spend and 89% of Claude Code spend. Hermes Agent spent the least on Claude at 1%.
Why it mattersModel choice drives most of a coding agent bill. Agents that support many models show where teams move when they have the choice.
Period
Oct 2026
Updated
November 11, 2026
ID
coding-agent-model-oct26
§ 01
Key findings
- 01Qwen Code: 92% Claude, 0% OpenAI.
- 02Claude Code: 89% Claude, 1% OpenAI.
- 03Zed: 76% Claude, 18% OpenAI.
- 04Zoo Code: 57% Claude, 7% OpenAI.
- 05OpenCode: 46% Claude, 24% OpenAI.
§ 01b
Questions this answers
- Which models do coding agents use most?
- How much coding agent spend goes to Claude?
- Which coding agents are the most model diverse?
§ 02
Data
| Agent | Claude share(percent) |
|---|---|
| Qwen Code | 91.76% |
| Claude Code | 88.84% |
| Zed | 75.70% |
| Zoo Code | 56.66% |
| OpenCode | 45.56% |
| Kilo Code | 43.02% |
| Roo Code | 32.21% |
| Cursor | 31.27% |
| GitHub Copilot | 30.70% |
| Pi | 16.10% |
| Cline | 3.27% |
| Hermes Agent | 0.57% |
§ 03
Cite as
APA
Click to copy
BibTeX
Click to copy
ID: coding-agent-model-oct26·Updated November 11, 2026·Period Oct 2026