@tuned-tensor/cli 0.4.18 → 0.4.20

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
@@ -34,6 +34,7 @@ tt init --name "Customer Support Bot" --model Qwen/Qwen3.5-2B
34
34
  # Edit tunedtensor.json, then:
35
35
  tt eval
36
36
  tt push
37
+ tt runs estimate <spec-id>
37
38
  tt runs start <spec-id>
38
39
  tt runs watch <run-id>
39
40
  ```
@@ -55,6 +56,20 @@ tt models base
55
56
  tt balance
56
57
  ```
57
58
 
59
+ Label real, unlabeled data with a teacher model (JSONL with `{"input": ...}`
60
+ rows, or CSV with `--input-column`; up to 50,000 rows / 50 MB). Labeling runs
61
+ as a managed cloud workflow — upload and disconnect; the teacher drafts
62
+ outputs under your spec's system prompt and you review before anything trains:
63
+
64
+ ```bash
65
+ tt label upload tickets.csv --spec <spec-id> --input-column body --watch
66
+ tt label watch <job-id> # re-attach to progress any time
67
+ tt label rows <job-id> --status labeled # review the teacher's drafts
68
+ tt label accept <job-id> --all # or accept/reject/edit by row
69
+ tt label promote <job-id> --name tickets-v1 # becomes a validated dataset
70
+ tt runs start <spec-id> --dataset <dataset-id>
71
+ ```
72
+
58
73
  Export a model to GGUF and package it for Ollama (so it's pluggable like any
59
74
  other local model, e.g. in OpenClaw via Ollama's native `/api/chat`):
60
75
 
@@ -72,8 +87,9 @@ This wraps llama.cpp's `convert_hf_to_gguf.py` + `llama-quantize` and Ollama's
72
87
  system prompt is embedded as the Modelfile `SYSTEM` block.
73
88
 
74
89
  For the full command reference, including dataset-backed runs, long-example
75
- policies, eval token budgets, continued fine-tuning, evaluation caps, local model serving,
76
- configuration, and billing, see the [CLI docs](https://tunedtensor.com/docs/cli).
90
+ policies, eval token budgets, preflight run estimates, continued fine-tuning,
91
+ evaluation caps, local model serving, configuration, and billing, see the
92
+ [CLI docs](https://tunedtensor.com/docs/cli).
77
93
 
78
94
  ## Development
79
95