Upload folder using huggingface_hub

README.md

@@ -99,8 +99,8 @@ for await (const text of stream) Deno.stdout.write(new TextEncoder().encode(text
 console.log(await stream.done); // { reason: "eos" | "max-tokens" | "aborted", … }
 ```
 
-Messages are plain `system` / `user` / `assistant` turns; tool calls, thinking channels
-and image or audio parts are rejected rather than silently dropped.
+Messages are plain `system` / `developer` / `user` / `assistant` turns; tool calls,
+thinking channels and image or audio parts are rejected rather than silently dropped.
 With no `sampler` in the request, generation uses this repository's recommended default: temperature 1.0, top-k 64, top-p 0.95.
 Weights are fetched once and cached (verified against `karume.json`'s `size` / `sha256`).
 
@@ -129,4 +129,4 @@ checked against each other when this repository was assembled.
 - **recommended sampler**: temperature 1.0, top-k 64, top-p 0.95 — used when a request omits
   `sampler`; pass `{ temperature: 0 }` for greedy decoding
 - **stop tokens** come from the tokenizer asset in this repository, not from the caller
-- **device limits**: `maxBufferSize` ≥ 402,653,184 B / `maxStorageBufferBindingSize` ≥ 402,653,184 B / `maxBufferSize` ≥ 402,653,184 B / `maxStorageBufferBindingSize` ≥ 402,653,184 B / `maxBufferSize` ≥ 402,653,184 B / `maxStorageBufferBindingSize` ≥ 402,653,184 B (the largest single tensor must bind)
+- **device limits**: `maxBufferSize` ≥ 402,653,184 B / `maxStorageBufferBindingSize` ≥ 402,653,184 B (the largest single tensor must bind)