You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: HISTORY.md
+18Lines changed: 18 additions & 0 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -1,5 +1,23 @@
1
1
# Release history
2
2
3
+
### main branch
4
+
5
+
- Added support for Claude 4.5/4.6 models and updated model aliases (sonnet/haiku/opus).
6
+
- Expanded Gemini model support with 2.5 Flash and Flash‑Lite, added Gemini 3 preview models, and updated the flash alias to gemini/gemini-flash-latest.
7
+
- Added DeepSeek Reasoner model and updated DeepSeek model metadata with costs and prompt caching.
8
+
- Added BadGatewayError and ImageFetchError to handled exceptions.
9
+
- Fixed exception mapping to only include real exception classes, avoiding runtime errors, by Claudia Pellegrino.
10
+
- Repo map now tags Fortran, Haskell, Julia, and Zig and is compatible with newer tree-sitter Python APIs, by Varchas Gopalaswamy, Tim Put, and Mubashir Osmani.
11
+
- Removed deprecated google-generativeai dependency to eliminate import warnings, by Mark McDonald.
12
+
- Added settings for new OpenAI GPT‑5.1/5.2 and GPT‑5‑pro models across OpenAI, Azure, and OpenRouter, including chat and codex variants.
13
+
- Changed the gemini alias to point to gemini/gemini-3-pro-preview.
14
+
- Added support for OpenAI o1-pro model.
15
+
- Aider wrote 57% of the code in this release.
16
+
17
+
### Aider v0.86.1
18
+
19
+
- Added a reasoning_effort setting for GPT-5 models and disabled temperature by default for the GPT-5 family.
20
+
3
21
### Aider v0.86.0
4
22
5
23
- Expanded GPT-5 model support across family variants and providers (OpenAI, Azure, OpenRouter), including dated and chat/mini/nano variants.
0 commit comments