diff --git a/core/src/browser/extensions/engines/helpers/sse.ts b/core/src/browser/extensions/engines/helpers/sse.ts
index bd9945d3cc..0842675825 100644
--- a/core/src/browser/extensions/engines/helpers/sse.ts
+++ b/core/src/browser/extensions/engines/helpers/sse.ts
@@ -38,7 +38,7 @@ export function requestInference(
errorCode = ErrorCode.InvalidApiKey
}
const error = {
- message: data.error?.message ?? 'Error occurred.',
+ message: data.error?.message ?? data.message ?? 'Error occurred.',
code: errorCode,
}
subscriber.error(error)
diff --git a/extensions/inference-cohere-extension/src/index.ts b/extensions/inference-cohere-extension/src/index.ts
index dd7f033174..2615ea8932 100644
--- a/extensions/inference-cohere-extension/src/index.ts
+++ b/extensions/inference-cohere-extension/src/index.ts
@@ -113,6 +113,8 @@ export default class JanInferenceCohereExtension extends RemoteOAIEngine {
}
transformResponse = (data: any) => {
- return typeof data === 'object' ? data.text : JSON.parse(data).text ?? ''
+ return typeof data === 'object'
+ ? data.text
+ : (JSON.parse(data.replace('data: ', '').trim()).text ?? '')
}
}
diff --git a/extensions/inference-openai-extension/package.json b/extensions/inference-openai-extension/package.json
index 50fe123493..9700383d63 100644
--- a/extensions/inference-openai-extension/package.json
+++ b/extensions/inference-openai-extension/package.json
@@ -1,7 +1,7 @@
{
"name": "@janhq/inference-openai-extension",
"productName": "OpenAI Inference Engine",
- "version": "1.0.3",
+ "version": "1.0.4",
"description": "This extension enables OpenAI chat completion API calls",
"main": "dist/index.js",
"module": "dist/module.js",
diff --git a/extensions/inference-openai-extension/resources/models.json b/extensions/inference-openai-extension/resources/models.json
index 3f41c0a7db..a34bc54604 100644
--- a/extensions/inference-openai-extension/resources/models.json
+++ b/extensions/inference-openai-extension/resources/models.json
@@ -99,6 +99,7 @@
"parameters": {
"temperature": 1,
"top_p": 1,
+ "stream": true,
"max_tokens": 32768,
"frequency_penalty": 0,
"presence_penalty": 0
@@ -126,6 +127,7 @@
"temperature": 1,
"top_p": 1,
"max_tokens": 65536,
+ "stream": true,
"frequency_penalty": 0,
"presence_penalty": 0
},
diff --git a/extensions/inference-openai-extension/src/index.ts b/extensions/inference-openai-extension/src/index.ts
index d484c8ae5e..18bc4e0aae 100644
--- a/extensions/inference-openai-extension/src/index.ts
+++ b/extensions/inference-openai-extension/src/index.ts
@@ -80,7 +80,6 @@ export default class JanInferenceOpenAIExtension extends RemoteOAIEngine {
return {
...params,
max_completion_tokens: max_tokens,
- stream: false, // o1 only support stream = false
}
}
// Pass through for non-preview models
diff --git a/web/containers/ErrorMessage/index.test.tsx b/web/containers/ErrorMessage/index.test.tsx
index d2ae5aa813..306a80e324 100644
--- a/web/containers/ErrorMessage/index.test.tsx
+++ b/web/containers/ErrorMessage/index.test.tsx
@@ -63,9 +63,6 @@ describe('ErrorMessage Component', () => {
render(
{message.content[0]?.text?.value && (