llm-stream-assemble 1.3.5 → 1.3.6
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +10 -4
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -1,11 +1,11 @@
|
|
|
1
1
|
# llm-stream-assemble
|
|
2
2
|
|
|
3
|
-
|
|
4
4
|
|
|
5
5
|
|
|
6
|
-
|
|
7
7
|
[](https://github.com/01laky/llm-stream-assemble/actions/workflows/ci.yml)
|
|
8
|
-
|
|
9
9
|
|
|
10
10
|
**One typed event model for every LLM stream** — text, tool calls, reasoning, JSON, usage, refusals, errors, and non-streaming responses.
|
|
11
11
|
|
|
@@ -13,7 +13,7 @@
|
|
|
13
13
|
|
|
14
14
|
Turn provider SSE fragments into typed events — **not another `+=` loop**.
|
|
15
15
|
|
|
16
|
-
**Status:** Stable `1.3.
|
|
16
|
+
**Status:** Stable `1.3.6`. Five built-in adapters, thirteen OpenAI-compatible host presets (including **Azure OpenAI** and **Cloudflare Workers AI**), transforms, replay helpers, and examples are production-ready. Pin semver ranges as usual and review [CHANGELOG.md](./CHANGELOG.md) before major upgrades.
|
|
17
17
|
|
|
18
18
|
---
|
|
19
19
|
|
|
@@ -31,6 +31,7 @@ Turn provider SSE fragments into typed events — **not another `+=` loop**.
|
|
|
31
31
|
- [Documentation](#documentation)
|
|
32
32
|
- [How this compares](#how-this-compares)
|
|
33
33
|
- [Examples](#examples)
|
|
34
|
+
- [Integration cookbook](#integration-cookbook)
|
|
34
35
|
- [Usage guides](#usage-guides)
|
|
35
36
|
- [Transforms & replay](#transforms--replay)
|
|
36
37
|
- [Examples & proxy safety](#examples--proxy-safety)
|
|
@@ -212,6 +213,7 @@ Pick an adapter in ~30 seconds:
|
|
|
212
213
|
- [Adapter author guide](./docs/adapter-guide.md)
|
|
213
214
|
- [Performance & runtime behavior](./docs/performance.md)
|
|
214
215
|
- [Edge-case showcase](./docs/edge-cases.md)
|
|
216
|
+
- [Integration cookbook](./docs/integration-cookbook.md)
|
|
215
217
|
- [How this compares](./docs/comparison.md)
|
|
216
218
|
- [FAQ](./docs/faq.md)
|
|
217
219
|
- [Architecture diagrams](./docs/img/README.md)
|
|
@@ -299,6 +301,10 @@ Stream `text.delta` into your renderer — this library does **not** parse markd
|
|
|
299
301
|
|
|
300
302
|
→ [`examples/node-fetch/replay-fixture.ts`](./examples/node-fetch/replay-fixture.ts)
|
|
301
303
|
|
|
304
|
+
### Integration cookbook
|
|
305
|
+
|
|
306
|
+
Wire unified events into **Hono**, **Express**, **Cloudflare Workers**, **LiteLLM**, **Next.js App Router**, AI SDK mapping, and LangChain callbacks — [`examples/integrations/`](./examples/integrations/) · **[Full cookbook →](./docs/integration-cookbook.md)**
|
|
307
|
+
|
|
302
308
|
---
|
|
303
309
|
|
|
304
310
|
## Usage guides
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "llm-stream-assemble",
|
|
3
|
-
"version": "1.3.
|
|
3
|
+
"version": "1.3.6",
|
|
4
4
|
"description": "Zero-dependency TypeScript stream assembly for OpenAI, Anthropic, Gemini, and OpenAI-compatible LLM APIs — unified StreamEvent model for text, tools, reasoning, JSON, usage, and non-streaming responses.",
|
|
5
5
|
"license": "MIT",
|
|
6
6
|
"author": "Ladislav Kostolny <01laky@gmail.com>",
|