@ax-llm/ax 21.0.9 → 21.0.11

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 CHANGED
@@ -45,6 +45,21 @@ providers with one line. No rewrites needed.
45
45
  Stop tweaking prompts. Define inputs → outputs. The framework generates optimal
46
46
  prompts automatically.
47
47
 
48
+ ### ⚡ **Provider-Native Speed**
49
+
50
+ Ax stays in the same latency class as direct provider calls while adding typed
51
+ outputs, validation, retries, tracing, and tools. Streaming is the default so Ax
52
+ can parse fields as they arrive, fail early on streaming assertions, cancel bad
53
+ outputs, and start correction without wasting tokens on completions that are
54
+ already invalid.
55
+
56
+ Benchmark the current code against your own providers:
57
+
58
+ ```bash
59
+ AX_STREAM_BENCH_PROVIDER=anthropic AX_STREAM_BENCH_MODEL=claude-sonnet-4-5-20250929 AX_STREAM_BENCH_RUNS=2 AX_STREAM_BENCH_WARMUP_RUNS=0 npm run tsx src/examples/streaming-latency.ts
60
+ AX_STREAM_BENCH_PROVIDER=google-gemini AX_STREAM_BENCH_MODEL=gemini-2.5-flash AX_STREAM_BENCH_RUNS=2 AX_STREAM_BENCH_WARMUP_RUNS=0 npm run tsx src/examples/streaming-latency.ts
61
+ ```
62
+
48
63
  ### 🛡️ **Production-Ready from Day One**
49
64
 
50
65
  Built-in streaming, validation, error handling, observability. Used by startups