@tuned-tensor/cli 0.13.1-beta.4 → 0.13.1-beta.5
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 +5 -2
- package/dist/{chunk-7VIUHQBI.js → chunk-AVMRHDFY.js} +1499 -792
- package/dist/chunk-AVMRHDFY.js.map +1 -0
- package/dist/index.js +155 -18
- package/dist/index.js.map +1 -1
- package/dist/local-runtime.js +31 -1
- package/dist/local-runtime.js.map +1 -1
- package/docs/local-runtime/spark.md +6 -0
- package/package.json +1 -1
- package/dist/chunk-7VIUHQBI.js.map +0 -1
|
@@ -15,11 +15,17 @@ certifies `Qwen/Qwen3.5-2B` at snapshot
|
|
|
15
15
|
Run on the Spark:
|
|
16
16
|
|
|
17
17
|
```bash
|
|
18
|
+
tt hardware
|
|
18
19
|
nvidia-smi
|
|
19
20
|
node --version
|
|
20
21
|
uv --version
|
|
21
22
|
```
|
|
22
23
|
|
|
24
|
+
`tt hardware` sizes certified adapter models and the foundation engine against
|
|
25
|
+
this GPU. Spark unified memory should show Nemotron and Muse Glimmer as
|
|
26
|
+
trainable; consumer cards typically only fit Qwen LoRA. `tt status` and the
|
|
27
|
+
shell agent reuse the cached report.
|
|
28
|
+
|
|
23
29
|
TT Local requires Node 22+, `uv`, working CUDA PyTorch, and enough free space
|
|
24
30
|
for the Hugging Face cache plus run artifacts.
|
|
25
31
|
|