From c883a87c8ecadedee9bc54f14c7a71d8d3ecfd82 Mon Sep 17 00:00:00 2001 From: Anshul Basia <73226526+anshulbasia27@users.noreply.github.com> Date: Tue, 17 Mar 2026 17:27:14 +0530 Subject: [PATCH 1/5] Update integrations/llms/ollama.mdx --- integrations/llms/ollama.mdx | 21 ++++++++++++++++++++- 1 file changed, 20 insertions(+), 1 deletion(-) diff --git a/integrations/llms/ollama.mdx b/integrations/llms/ollama.mdx index f9117eaf..b7710d71 100644 --- a/integrations/llms/ollama.mdx +++ b/integrations/llms/ollama.mdx @@ -22,11 +22,13 @@ docker run -d -p 8787:8787 portkeyai/gateway:latest Then, connect to your local Ollama instance: + + ```python Python from portkey_ai import Portkey portkey = Portkey( - base_url="http://localhost:8787", # Your local Gateway + base_url="http://localhost:8787/v1", # Your local Gateway provider="ollama", custom_host="http://localhost:11434" # Your Ollama instance ) @@ -37,6 +39,23 @@ response = portkey.chat.completions.create( ) ``` +```javascript Node.js +import Portkey from 'portkey-ai'; + +const portkey = new Portkey({ + baseURL: 'http://localhost:8787/v1', // Your local Gateway + provider: 'ollama', + customHost: 'http://localhost:11434' // Your Ollama instance +}); + +const response = await portkey.chat.completions.create({ + model: 'llama3', + messages: [{ role: 'user', content: 'Hello!' }] +}); +``` + + + From 9047d5d24418d35c737a1d126017ac9e7f6aaa62 Mon Sep 17 00:00:00 2001 From: Anshul Basia <73226526+anshulbasia27@users.noreply.github.com> Date: Tue, 17 Mar 2026 17:41:44 +0530 Subject: [PATCH 2/5] docs: clarify that base_url requires /v1 for local gateway setup --- integrations/llms/ollama.mdx | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/integrations/llms/ollama.mdx b/integrations/llms/ollama.mdx index b7710d71..aefd912e 100644 --- a/integrations/llms/ollama.mdx +++ b/integrations/llms/ollama.mdx @@ -153,7 +153,7 @@ const portkey = new Portkey({ -**Important:** For Ollama integration, you only need to pass the base URL to `customHost` **without** the version identifier (such as `/v1`) - Portkey handles the rest! +**Important:** The `custom_host` / `customHost` parameter (your Ollama URL) should be passed **without** `/v1` — Portkey handles the provider routing automatically. However, when using a local gateway, the `base_url` / `baseURL` parameter **must** include `/v1` (e.g. `http://localhost:8787/v1`). --- From 83ca538eaba132c8675dc30c145e0151cdeb435d Mon Sep 17 00:00:00 2001 From: Akhil Madhu Menon Date: Fri, 3 Apr 2026 14:21:41 +0530 Subject: [PATCH 3/5] Added API Key reset usage example --- .../control-plane/api-keys/update-api-key.mdx | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/api-reference/admin-api/control-plane/api-keys/update-api-key.mdx b/api-reference/admin-api/control-plane/api-keys/update-api-key.mdx index bfbc1b8b..b8feb04a 100644 --- a/api-reference/admin-api/control-plane/api-keys/update-api-key.mdx +++ b/api-reference/admin-api/control-plane/api-keys/update-api-key.mdx @@ -2,3 +2,15 @@ title: Update API Key openapi: put /api-keys/{id} --- + +### Reset API key usage + +Use `reset_usage` when you want to clear the API key's current usage without changing other fields. + +```json +{ + "reset_usage": true +} +``` + +If the key is currently `exhausted`, this reset reactivates it to `active` (when applicable). After the update, confirm the key reflects reset behavior in response fields such as `status` and `last_reset_at`. From 0d05e5666ce38a4fd331ec99d198cd1292eb486d Mon Sep 17 00:00:00 2001 From: Akhil Madhu Menon Date: Fri, 3 Apr 2026 14:42:11 +0530 Subject: [PATCH 4/5] Reverting changes to ollama.mdx --- integrations/llms/ollama.mdx | 25 +++---------------------- 1 file changed, 3 insertions(+), 22 deletions(-) diff --git a/integrations/llms/ollama.mdx b/integrations/llms/ollama.mdx index aefd912e..ef1896e8 100644 --- a/integrations/llms/ollama.mdx +++ b/integrations/llms/ollama.mdx @@ -22,13 +22,11 @@ docker run -d -p 8787:8787 portkeyai/gateway:latest Then, connect to your local Ollama instance: - - ```python Python from portkey_ai import Portkey portkey = Portkey( - base_url="http://localhost:8787/v1", # Your local Gateway + base_url="http://localhost:8787", # Your local Gateway provider="ollama", custom_host="http://localhost:11434" # Your Ollama instance ) @@ -39,23 +37,6 @@ response = portkey.chat.completions.create( ) ``` -```javascript Node.js -import Portkey from 'portkey-ai'; - -const portkey = new Portkey({ - baseURL: 'http://localhost:8787/v1', // Your local Gateway - provider: 'ollama', - customHost: 'http://localhost:11434' // Your Ollama instance -}); - -const response = await portkey.chat.completions.create({ - model: 'llama3', - messages: [{ role: 'user', content: 'Hello!' }] -}); -``` - - - @@ -153,7 +134,7 @@ const portkey = new Portkey({ -**Important:** The `custom_host` / `customHost` parameter (your Ollama URL) should be passed **without** `/v1` — Portkey handles the provider routing automatically. However, when using a local gateway, the `base_url` / `baseURL` parameter **must** include `/v1` (e.g. `http://localhost:8787/v1`). +**Important:** For Ollama integration, you only need to pass the base URL to `customHost` **without** the version identifier (such as `/v1`) - Portkey handles the rest! --- @@ -194,4 +175,4 @@ For complete SDK documentation: Complete Portkey SDK documentation - + \ No newline at end of file From f6eb0f3f81d225ae26c703229ded9a54ad1a86a1 Mon Sep 17 00:00:00 2001 From: Akhil Madhu Menon Date: Fri, 3 Apr 2026 14:46:29 +0530 Subject: [PATCH 5/5] Another attempt to revert this file --- integrations/llms/ollama.mdx | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/integrations/llms/ollama.mdx b/integrations/llms/ollama.mdx index ef1896e8..f9117eaf 100644 --- a/integrations/llms/ollama.mdx +++ b/integrations/llms/ollama.mdx @@ -175,4 +175,4 @@ For complete SDK documentation: Complete Portkey SDK documentation - \ No newline at end of file +