041b37b64f
switch gpt-5 to default to codex prompt + high reasoning
2025-09-01 21:46:03 -04:00
aea78046d0
Refactor to support multiple instances inside single opencode process ( #2360 )
...
This release has a bunch of minor breaking changes if you are using opencode plugins or sdk
1. storage events have been removed (we might bring this back but had some issues)
2. concept of `app` is gone - there is a new concept called `project` and endpoints to list projects and get the current project
3. plugin receives `directory` which is cwd and `worktree` which is where the root of the project is if it's a git repo
4. the session.chat function has been renamed to session.prompt in sdk. it no longer requires model to be passed in (model is now an object)
5. every endpoint takes an optional `directory` parameter to operate as though opencode is running in that directory
2025-09-01 17:15:49 -04:00
bf77aa6acd
feat: Allow provider timeout override ( #1982 )
2025-08-31 14:06:02 -04:00
1bd97ef802
slash commands ( #2157 )
...
Co-authored-by: adamdotdevin <2363879+adamdottv@users.noreply.github.com >
2025-08-22 17:04:28 -04:00
f69816302b
sonic model
2025-08-20 01:02:41 -04:00
7ed4a97ad2
blacklist gpt-5-chat-latest ( #2048 )
2025-08-18 17:50:38 -04:00
ba30792e43
allow plugins to create custom auth providers
2025-08-14 16:25:08 -04:00
0d7f3dedeb
fix: reasoning not supported ( #1882 )
2025-08-13 06:26:07 -05:00
fea9cc399a
add openai prompt cache key
2025-08-12 17:37:15 -04:00
66d59549ac
azure reasoning effort
2025-08-11 23:04:32 -04:00
91a7b503d0
update anthropic prompt and variables
2025-08-11 21:01:33 -04:00
5ec4233910
ignore: fix share page
2025-08-11 16:12:26 -04:00
d3425b976f
fix azure gpt config
2025-08-11 10:56:16 -04:00
80347b1ad4
feat: thinking blocks rendered in tui and share page
2025-08-10 19:25:03 -05:00
b5798d2dae
use minimal reasoning effort for gpt-5
2025-08-09 15:38:48 -04:00
e530a3287a
gpt-5 lower verbosity
2025-08-08 20:42:22 -04:00
56587b1849
temporary fix for max output token
2025-08-07 22:54:59 -04:00
e3781104c4
increase models.dev polling interval to hourly
2025-08-03 14:58:35 -04:00
d3ca086b59
wip: plugins
2025-08-02 18:50:19 -04:00
587ab4f16a
ignore: fix dev branch ( #1529 )
2025-08-02 09:11:38 -05:00
55a73b3e2e
feat(models): enable Kimi k2 ⇄ Claude trajectory handoff ( #1525 )
2025-08-01 23:05:06 -04:00
f75fc5af22
force models.dev refresh on auth login
2025-08-01 17:48:01 -04:00
cbd7db5a73
make top_p configurable
2025-08-01 17:03:33 -04:00
feb43e8550
add version to user-agent
2025-08-01 12:18:09 -04:00
c64cd54d5d
poll for models.dev changes
2025-07-31 23:47:42 -04:00
d944eed4d5
add user agent for models.dev request
2025-07-31 22:00:45 -04:00
b67c367a40
Use responses API for Azure ( #1428 )
publish / publish (push) Has been cancelled
2025-07-30 23:22:59 -04:00
30e5d32ed8
Add http-referer header for vercel ai gateway requests ( #1403 )
2025-07-30 23:22:24 -04:00
5a3aed4e88
qwen optimizations it works good now
2025-07-25 18:31:08 -04:00
4b94f390ce
allow temperature to be configured per mode
publish / publish (push) Has been cancelled
2025-07-25 13:29:44 -04:00
8e036cf17b
fix: apply content-level caching for non-anthropic providers ( #1305 )
2025-07-25 12:19:44 -04:00
0de957a2f0
real life totally configurabl ai subasians
publish / publish (push) Has been cancelled
2025-07-24 21:21:02 -04:00
5c1d6d3c8d
Check if modelID includes "claude" for antropic/claude prompt caching ( #1284 )
2025-07-24 11:31:28 -04:00
f5f0f57098
fix openrouter caching
publish / publish (push) Has been cancelled
2025-07-19 15:11:21 -04:00
c7ae049c93
pass through model.options properly without having to nest it under provider name. you may have to update your configs see https://opencode.ai/docs/models/#openrouter for an example
publish / publish (push) Has been cancelled
2025-07-19 12:41:58 -04:00
549e66fc8b
feat(provider): add Gemini tool schema sanitization ( #1132 )
2025-07-18 16:02:54 -04:00
aef44471dd
feat(config): make small model configurable ( #1030 )
publish / publish (push) Has been cancelled
2025-07-18 14:16:50 -04:00
d45c1a4980
fix: model cost overrides
publish / publish (push) Has been cancelled
2025-07-18 05:08:35 -05:00
d09f2f72e5
feat: support AWS_BEARER_TOKEN_BEDROCK for amazon bedrock provider autoloading ( #1094 )
publish / publish (push) Has been cancelled
2025-07-17 09:12:30 -05:00
7d7ed19879
add small model for title generation
publish / publish (push) Has been cancelled
2025-07-15 14:00:52 -04:00
f72567716f
bring back task tool
publish / publish (push) Has been cancelled
2025-07-15 00:05:54 -04:00
c74a8a566d
fix: avoid overwriting the provider.option.baseURL ( #880 )
2025-07-11 18:01:28 -04:00
1dd8a8c48a
fix(copilot): add vision request header ( #773 )
2025-07-08 14:01:54 -05:00
9cad702612
v2 message format and upgrade to ai sdk v5 ( #743 )
...
publish / publish (push) Has been cancelled
Co-authored-by: GitHub Action <action@github.com >
Co-authored-by: Liang-Shih Lin <liangshihlin@proton.me >
Co-authored-by: Dominik Engelhardt <dominikengelhardt@ymail.com >
Co-authored-by: Jay V <air@live.ca >
Co-authored-by: adamdottv <2363879+adamdottv@users.noreply.github.com >
2025-07-07 15:53:43 -04:00
207c73bc6a
fix openrouter caching with anthropic, should be a lot cheaper
2025-07-05 11:39:54 -04:00
e9e786746f
fix: typescript error
2025-07-02 16:08:41 -05:00
1dd3bcae9f
Changed handling of OpenRouter requests to add some custom headers so that it can see the app ( #613 )
...
Co-authored-by: Dax Raad <d@ironbay.co >
2025-07-02 14:43:59 -04:00
70dbfa4d00
fix: Copilot Premium Requests ( #595 )
2025-07-02 12:04:53 -04:00
e8c1c74b94
fix loading api key from env for openai compatible providers
publish / publish (push) Has been cancelled
2025-06-30 19:07:51 -04:00
de2142c9a3
chore: typescript error
2025-06-30 15:46:18 -05:00
Dax Raad