		{
		"name": "@axflow/models",
		"version": "0.0.23",
		"version": "0.0.24",
		"description": "Zero-dependency, modular SDK for building robust natural language applications",
		@@ -113,2 +113,5 @@ "author": "Axflow (https://axflow.dev)",
		],
		"togetherai": [
		"./dist/togetherai/inference.d.ts"
		],
		"shared": [
		@@ -181,2 +184,8 @@ "./dist/shared/index.d.ts"
		},
		"./togetherai/inference": {
		"types": "./dist/togetherai/inference.d.ts",
		"import": "./dist/togetherai/inference.mjs",
		"module": "./dist/togetherai/inference.mjs",
		"require": "./dist/togetherai/inference.js"
		},
		"./react": {
		@@ -201,3 +210,3 @@ "types": "./dist/react/index.d.ts",
		},
		"gitHead": "3f3abc72d4e38a4201d0425f0b8ae29fa283725c"
		"gitHead": "ae40ef75896a60c79c3a0527a3b3371a15935ec2"
		}

README.md

		@@ -11,2 +11,3 @@ # @axflow/models

		- Support for all leading model providers, plus popular open source models like Llama2 or Mistral through Together.ai inference
		- First-class streaming support for both low-level byte streams and higher-level JavaScript object streams
		@@ -27,2 +28,3 @@ - First-class support for streaming arbitrary data in addition to the LLM response
		- ✅ Azure OpenAI
		- ✅ TogetherAI (Llama2, Mistral, etc.)
		- Google PaLM models (coming soon)
		@@ -45,2 +47,3 @@ - Replicate (coming soon)
		- [@axflow/models/ollama/embedding](https://docs.axflow.dev/documentation/models/ollama-embedding.html)
		- [@axflow/models/togetherai/inference](https://docs.axflow.dev/documentation/models/togetherai-inference.html)
		- [@axflow/models/react](https://docs.axflow.dev/documentation/models/react.html)
		@@ -47,0 +50,0 @@ - [@axflow/models/node](https://docs.axflow.dev/documentation/models/node.html)

@axflow/models - npm Package Compare versions

New alerts