Commit Graph
891 Commits
Author SHA1 Message Date
Rohit C Prasad fc6ca8a703 Custom providers: permanent add box, named cards, fetch models from server
The box stays at the end of the API-keys group; saving makes a card titled by the user, prefix derived from the name.
A custom card lists its server's /models on request (a missing listing is a note, not a failure).
Re-saving a custom key now rebuilds its client; strings in en + zh.
2026-10-26 11:16:46 -07:00
Rohit C Prasad b849b53d27 Add Cerebras, Hugging Face, Cohere and SiliconFlow as compat providers
One verified tool-calling row each, with the window from the vendor's docs (2026-10-26).
SiliconFlow's endpoint help names the China mainland variant.
2026-10-26 11:10:57 -07:00
Rohit C Prasad db6e0af7eb Add Jan, Docker Model Runner and mlx-lm as local server providers
Same shape as llama.cpp/vLLM/LM Studio: address + optional key, models from the server.
Docker's /engines path is added when the user types the root; mlx-lm's help covers the port shared with llama.cpp.
Local-server checks now read LOCAL_SERVERS instead of a hand-kept tuple.
2026-10-26 11:10:29 -07:00
Mumit Khan e9777cbb1b Add custom OpenAI-compatible providers 2026-10-26 11:01:42 -07:00
Rohit Prasad 99ee5e000c Merge pull request #784 from andrewyng/feat/lmstudio-local-server
LM Studio as a local server provider
2026-10-25 21:12:51 -07:00
Rohit Prasad 7eceed6f6b Merge pull request #59 from hunkim/add-upstage-provider
Add Upstage (Solar) as an OpenAI-compatible model provider
2026-10-25 21:03:21 -07:00
Rohit C Prasad c07f8096a3 Screenshots of the LM Studio provider
Light and dark, for the PR and the docs.
2026-10-25 20:55:44 -07:00
Rohit C Prasad 45d58dc9a5 Add LM Studio as a local server provider
Joins llama.cpp and vLLM: address + optional key, window from its own listing.
Based on #434 by zalex666: name, model and help text.
2026-10-25 20:55:44 -07:00
Rohit C Prasad 66c91283eb Upstage: Solar Pro 4 and Mini 4 aliases with windows, tail of the gallery
solar-open2 and solar-pro3 are retired; the two current aliases carry the 512K the docs list (2026-10-25).
2026-10-25 20:55:16 -07:00
Sung Kim bd99ef3af9 fix(upstage.svg): correct gradient ID typo (upsate → upstage) and add trailing newline 2026-10-25 20:55:16 -07:00
Sung KimandClaude Fable 5 6e00f42006 Make solar-open2 the recommended Upstage model
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-10-25 20:55:16 -07:00
Sung KimandClaude Fable 5 d8ac7cefcf Add Upstage (Solar) as an OpenAI-compatible model provider
Wires Upstage into the existing compat-vendor pattern: a _compat descriptor
(prefilled https://api.upstage.ai/v1 endpoint, UPSTAGE_API_KEY env fallback),
a solar-pro3 matrix entry, the "solar" capability heuristic, suggested-model
extras, and the GUI key-help link + vendored lobe-icons logo.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-10-25 20:55:16 -07:00
Rohit Prasad 6e7ced9d61 Merge pull request #44 from BrightProgrammer7/add-groq-openrouter-providers
feat(providers): add Groq and OpenRouter as first-class resellers
2026-10-25 20:52:50 -07:00
Rohit Prasad 4d7a273903 Merge pull request #273 from jason-wu-ai/novita-provider-clean
Add Novita AI provider
2026-10-25 20:52:40 -07:00
Rohit Prasad 2480d44799 Merge pull request #668 from binyangzhu000-sudo/codex/atlascloud-provider
feat(providers): add Atlas Cloud
2026-10-25 20:52:31 -07:00
Rohit Prasad 4c4e537322 Merge pull request #185 from hrishinave/feat/nvidia-nim-provider
Add NVIDIA NIM provider support
2026-10-25 20:52:03 -07:00
Rohit Prasad e8433b1236 Merge pull request #503 from XiaoHuo888-hue/feat/orcarouter-provider
feat: add OrcaRouter as a named model provider
2026-10-25 20:51:54 -07:00
Rohit C Prasad 25b30125dc Groq: GPT-OSS rows take tools one at a time
Groq's tool-use page lists parallel tool use for Llama 3.3 only (2026-10-25).
2026-10-25 20:41:46 -07:00
Rohit C Prasad 4f6665817a Groq: current production rows with windows, key link, brand mark
Three ids from the models page with 131,072 windows (2026-10-25); the deferral note is closed.
2026-10-25 20:41:46 -07:00
CleverCoder 24356cc950 feat(providers): add Groq and OpenRouter as first-class resellers
Resolves the TODO in matrix.py and registry.py — both providers use the
existing _compat helper (OpenAI-compatible endpoints).

Groq: 3 curated models (Llama 4 Maverick, DeepSeek R1 Distill 70B, Qwen QwQ 32B).
OpenRouter: 4 curated models spanning Anthropic, Google, DeepSeek, and Meta.

Tests extended: reseller lockstep and capability assertions now cover both.
2026-10-25 20:41:35 -07:00
Rohit C Prasad 5c3f4d71bf Novita: recommended GLM-5.2 row, key page link, brand mark
Keeps the factory's required recommended model; window 1,048,576 from the live catalogue (2026-10-25).
2026-10-25 20:41:14 -07:00
freecodewu fac0556f8e Add Novita AI provider 2026-10-25 20:40:52 -07:00
Rohit C Prasad 60ba5957c5 Atlas Cloud: GLM-5.2 as the served row, console key link, brand mark
Id and 1,048,576 window from the live catalogue (2026-10-25); gpt-4.1-mini is no longer served.
2026-10-25 20:40:28 -07:00
binyangzhu000-sudo 5d6107d6c5 feat(providers): add Atlas Cloud
Signed-off-by: binyangzhu000-sudo <224954946+binyangzhu000-sudo@users.noreply.github.com>
2026-10-25 20:40:28 -07:00
Rohit C Prasad e81daae2be SambaNova: three served rows, key link, brand mark, tail of the gallery
Rows and windows from the live catalogue (2026-10-25); the retired M2.7 row goes.
2026-10-25 20:40:04 -07:00
snova-kwasia 1f2f72f53b Add SambaNova as an OpenAI-compatible provider
- Added sambanova to registry.py via _compat() helper (OpenAI-compatible API)
- Added 6 SambaNova models to matrix.py (Llama 3.3 70B, DeepSeek V3.1/V3.2, MiniMax M2.7, Gemma 4 31B, GPT-OSS 120B)
- Uses official endpoint https://api.sambanova.ai/v1 with SAMBANOVA_API_KEY env var
2026-10-25 20:39:43 -07:00
Rohit C Prasad af45cbe492 NVIDIA NIM: current Nemotron ids with windows, self-hosted Test lists models
Three rows from the live catalogue with the 1M windows NVIDIA lists (2026-10-25).
A non-default endpoint is probed with GET /models, so Test works on a self-hosted NIM.
2026-10-25 20:39:21 -07:00
Hrishikesh Naveenam 2191743f23 Add NVIDIA NIM provider support 2026-10-25 20:39:21 -07:00
Rohit C Prasad ccf7ba9199 OrcaRouter: two rows verified live, console key link, tail of the gallery
Sonnet 5 (1,000,000) and DeepSeek V4 Pro (1,048,576) from the public catalogue, 2026-10-25.
Suggestions come from the matrix; the default endpoint help applies.
2026-10-25 20:38:57 -07:00
XiaoHuo888andClaude 38bbe717a3 feat: add OrcaRouter as a named model provider
Registers OrcaRouter (https://www.orcarouter.ai) as a first-class
OpenAI-compatible provider, mirroring the existing OpenRouter wiring:
prefilled endpoint, ORCAROUTER_API_KEY (sk-orca- keys), server + GUI key
detection, verify support, three curated matrix models (Claude Sonnet 5,
GPT-5.5, DeepSeek V4 Pro), a GUI logo and keys link, and tests.

Co-Authored-By: Claude <noreply@anthropic.com>
2026-10-25 20:38:41 -07:00
Rohit Prasad cda5d8a376 Merge pull request #193 from amrrs/nebius-token-factory
Add Nebius Token Factory provider
2026-10-25 20:37:22 -07:00
Rohit C Prasad 7a3c14ec21 Nebius: two verified rows with windows, tail of the gallery
Ids and 1,048,576 windows from the public catalogue (2026-10-25).
Drops the retired AI Studio clause; Nebius goes after xAI.
2026-10-25 19:48:59 -07:00
Rohit Prasad 4cbc1b1ccc Merge pull request #783 from andrewyng/feat/settings-customization
Settings: Customization tab for the web search service
2026-10-25 19:47:45 -07:00
amrrsandCursor 0bdd7b7b35 Add Nebius Token Factory provider
Nebius Token Factory serves many labs' models behind one key over an
OpenAI-compatible API, so it joins Together/Fireworks as a reseller
descriptor plus curated matrix rows in their own namespaces.

Verified against the live API on 2026-07-26: /v1/models, a tool-call
round-trip and streaming on all four curated ids, and verify_provider_key
for both a good and a bad key.

Co-authored-by: Cursor <cursoragent@cursor.com>
2026-10-25 19:45:45 -07:00
Rohit C Prasad aa9a6f9bba Screenshots of the Customization tab
Light and dark, for the PR and the docs.
2026-10-25 19:38:29 -07:00
Rohit C Prasad 7219e3a3c9 Customization: fetch option sits under the active provider
One indented checkbox under Exa, Firecrawl or Tavily when it is in use, with a caution line; the separate card is gone.
The web_fetch card line now reads "fetched by your search service, <name>".
2026-10-25 19:32:05 -07:00
Rohit C Prasad 301f6f7390 Tests follow the web search settings API
The grant-clearing test picks a keyless provider; the approval card test covers the fetch-provider line.
2026-10-25 19:21:12 -07:00
Rohit C Prasad c35b557a1e Approval card names the service that fetches the page
web_fetch carries fetch_provider when pages go through the search service; one gallery state for it.
2026-10-25 19:16:25 -07:00
Rohit C Prasad d1d9b12227 Settings: Customization tab with the Web search section
Providers as radio rows with a key per service (test and save), and the fetch-through-the-search-service switch.
Between Connectors and Context optimization; deep link #/settings/customization.
2026-10-25 19:16:25 -07:00
Rohit C Prasad 535467a4d2 Point the web_search_provider notes at Settings > Customization 2026-10-25 19:06:58 -07:00
Rohit C Prasad 3cc57ad0a0 web_fetch can read pages through Exa, Firecrawl or Tavily
Same address checks first, then the provider returns the page; a failure falls back to a local fetch with a notice.
The approval card learns which service fetches.
2026-10-25 19:06:58 -07:00
Rohit C Prasad 6e844c3579 Web search settings: one key per provider, test-and-save, fetch switch
The web_search:default profile keeps keys per provider; the legacy api_key shape is read as keys[provider].
Routes: GET/POST /v1/web-search, PUT/DELETE /v1/web-search/keys/{name}, POST /v1/web-search/fetch-via-provider.
2026-10-25 19:06:58 -07:00
Rohit Prasad 5214d76867 Merge pull request #262 from brainsparker/you-search-provider
Add You.com web search provider (keyless, with optional key)
2026-10-25 18:56:27 -07:00
Rohit Prasad fe6d4ba090 Merge pull request #86 from rakshith48/add-firecrawl-web-search
Add Firecrawl as a web_search provider
2026-10-25 18:56:17 -07:00
Rohit Prasad f1476e2978 Merge pull request #360 from kesku/kesku/add-exa-support
Add Exa web search provider
2026-10-25 18:56:06 -07:00
Rohit Prasad f4b569a759 Merge pull request #233 from manavshrivastavagit/serper-provider
Add Serper.dev search provider
2026-10-25 18:55:56 -07:00
Rohit C Prasad d55ca63097 Register You.com with the site wall and make key handling uniform
Both You.com hosts join PROVIDER_HOSTS (keyed vs keyless) and the Sites catalogue; the YDC_API_KEY fallback moves to key resolution.
Every provider declares key_mode none/optional/required beside display_name and key_url.
2026-10-25 16:28:18 -07:00
Rohit Prasad 5d8477e484 Merge pull request #781 from andrewyng/chore/matrix-cap-100
Model list cap: 75 to 100
2026-10-25 16:27:36 -07:00
Brian Sparker a30548f85c Add You.com web search provider
You.com is usable with no API key at all — the free agents endpoint gives
100 searches/day, rate-limited per IP — so it works on a fresh install the
way the ddgs default does, and switches to the Search API when a key is
configured.

Supporting that needed one change outside the provider class. Key handling
was a single boolean, and build_provider discarded the API key for any
provider with requires_key = False, so a provider that is keyless but
better with a key could not be expressed: a key set through the SecretStore
or POST /v1/web-search would be silently ignored. WebSearchProvider gains
an accepts_key flag, defaulting to False, and build_provider passes the key
through when it is set. DuckDuckGo, Tavily and Brave are unaffected.

Nothing else moves. SearchResult keeps its three fields, the manager, the
/v1/web-search endpoints and the agent registry all read provider_names(),
and resolve_provider() already derives YOU_API_KEY from the provider name
(YDC_API_KEY is accepted as a fallback, the name You.com's own docs use).
httpx is already a core dependency.

Requests send User-Agent: youdotcom-integration/andrewyng-openworker, to
You.com hosts only. On the keyless tier there is no key, so this is the
only way the traffic can be attributed.
2026-10-25 16:26:02 -07:00
Rohit C Prasad c59c288fe3 Register Firecrawl with the site wall
Firecrawl's host joins PROVIDER_HOSTS and the Access ▸ Sites catalogue.
The class gets display_name and key_url and raises on HTTP errors like the others.
2026-10-25 16:25:35 -07:00