soup-cli 0.40.0__tar.gz → 0.40.2__tar.gz

{soup_cli-0.40.0 → soup_cli-0.40.2}/CONTRIBUTING.md

@@ -111,7 +111,7 @@ soup_cli/
   templates/          - 17 built-in soup.yaml templates (YAML + manifest.json) with load_template loader (v0.39.0, +bco v0.40.0)
   ui/                 - Web UI (FastAPI + HTML/JS SPA)
 
-tests/                - Test suite (136 files, 4656 tests)
+tests/                - Test suite (143 files, 4756 tests)
 examples/             - Real-world config examples and datasets
 ```
 

{soup_cli-0.40.0 → soup_cli-0.40.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: soup-cli
-Version: 0.40.0
+Version: 0.40.2
 Summary: Fine-tune LLMs in one command. No SSH, no config hell.
 Project-URL: Homepage, https://github.com/MakazhanAlpamys/Soup
 Project-URL: Repository, https://github.com/MakazhanAlpamys/Soup
@@ -27,7 +27,7 @@ Requires-Dist: pydantic>=2.0.0
 Requires-Dist: pyyaml>=6.0
 Requires-Dist: rich>=13.0.0
 Requires-Dist: torch>=2.0.0
-Requires-Dist: transformers>=4.36.0
+Requires-Dist: transformers<5.0.0,>=4.36.0
 Requires-Dist: trl>=0.7.0
 Requires-Dist: typer<0.21.0,>=0.9.0
 Provides-Extra: audio
@@ -100,6 +100,7 @@ Description-Content-Type: text/markdown
 </p>
 
 <p align="center">
+  <a href="https://trysoup.dev">Website</a> &middot;
   <a href="#quick-start">Quick Start</a> &middot;
   <a href="#features">Features</a> &middot;
   <a href="#data-tools">Data Tools</a> &middot;
@@ -115,6 +116,7 @@ Description-Content-Type: text/markdown
   <img src="https://img.shields.io/badge/license-Apache--2.0-blue" alt="Apache-2.0 License">
   <a href="https://github.com/MakazhanAlpamys/Soup/actions"><img src="https://img.shields.io/endpoint?url=https://gist.githubusercontent.com/MakazhanAlpamys/65fdc943f85f3b2c46ecddb415c2b779/raw/soup_tests.json" alt="Tests"></a>
   <a href="https://github.com/MakazhanAlpamys/Soup/actions"><img src="https://github.com/MakazhanAlpamys/Soup/actions/workflows/ci.yml/badge.svg" alt="CI"></a>
+  <a href="https://trysoup.dev"><img src="https://img.shields.io/badge/website-trysoup.dev-blue" alt="Website"></a>
 </p>
 
 ---
@@ -129,15 +131,20 @@ soup train
 
 ## What's New
 
-Latest highlights only. Full history: [GitHub Releases](https://github.com/MakazhanAlpamys/Soup/releases).
 
-**v0.40.0 — Preference Variety**: BCO trainer + a unified preference-loss surface so DPO / SimPO / ORPO / IPO / BCO live behind one config knob. Adds two opt-in DPO controls (β-schedule + ref-model regen) and a forward-looking multi-objective preference-loss surface.
+Latest highlights only. Full history: [GitHub Releases](https://github.com/MakazhanAlpamys/Soup/releases).
 
-- **BCO Trainer** — set `task: bco` for Binary Classifier Optimization. Same input format as DPO (`prompt + chosen + rejected`); rows are internally split to TRL's BCO unpaired schema. New `training.bco_beta` (default 0.1, gt=0). Template: `soup init --template bco`.
-- **Unified preference dispatcher** — set `task: preference` + `training.preference_loss: dpo|simpo|orpo|ipo|bco` to pick the loss without renaming your task. Legacy `task: dpo`, `task: simpo`, etc remain first-class — the new surface is additive, not a breaking collapse. Useful for hyperparameter sweeps over the loss type itself.
-- **KL-controlled DPO variants** — anneal β over training with `training.dpo_beta_schedule: linear|cosine|exponential` + `training.dpo_beta_end`. Periodically refresh the frozen reference model with the current student via `training.dpo_ref_regen_epochs: 2`. Both gated to DPO-family tasks (`dpo`, `ipo`, or `preference` with `preference_loss in {dpo, ipo}`); transformers backend only.
-- **Multi-objective preference loss** — define `training.preference_loss_weights: {dpo: 0.7, bco: 0.3}` to blend losses. 2–5 entries, weights must sum to 1. Schema-level surface ships now; live runtime weighted-loss combination deferred to v0.40.1 (`PreferenceTrainerWrapper.setup` raises `NotImplementedError` with a friendly message until then — same stub-then-live pattern as v0.27.0 MII / v0.37.0 multipack / v0.38.0 quant menu / v0.39.0 ReLoRA).
-- **Net +118 tests** (4538 → 4656) across BCO trainer + dispatcher + β schedule math + ref-model regen TOCTOU + multi-objective schema bounds.
+**v0.40.2 — Quick polish + carry-overs**: closes 3 originally-scheduled GitHub issues (#36 eval-gate dashboard row, #50 smarter `--hf-resume`, #51 custom HF Space templates) plus 7 v0.40.1 long-tail UX papercuts.
+
+- **`--hf-resume` prefers local newer** — `prepare_hf_resume` checks the highest local `checkpoint-N` against the highest remote branch and skips the snapshot download when the local copy is newer or equal. Saves bandwidth and never overwrites a fresher local checkpoint with stale Hub state.
+- **Eval-gate dashboard row** — pure formatter `format_gate_row(state)` lives in `soup_cli/monitoring/display.py` and renders a one-liner like `Gate: helpfulness 7.8 ✓ | math 0.82 ✗ (-0.06) | STOP` for the live training panel. Hidden when eval-gate is disabled.
+- **`soup deploy hf-space --template-dir <path>`** — supply your own `app.py` + `README.md` (+ optional `requirements.txt`) instead of the built-in `gradio-chat` / `streamlit-chat`. Containment-checked, repo-id substitution validated *before* render, 256 KB cap per file, symlinks rejected.
+- **`soup quickstart --output DIR`** — route data, config, and run dir under any directory you choose (default keeps backwards-compat in cwd).
+- **`soup runs --cwd-only`** — restrict the listing to runs whose `output_dir` is under the current directory; the global `~/.soup/experiments.db` view is still default.
+- **`soup infer` / `soup bench` accept HF ids** — when the local model path is missing AND the value isn't path-like (no `./`, `/`, `~`, `C:\`), it falls through to a HuggingFace download via `transformers.from_pretrained`. Path-like-but-missing surfaces a friendly `FileNotFoundError`.
+- **CLI flag aliases** — `data filter --min-coherence` (alias for `--coherence`); `data split --train` accepted (informational, train is the implicit remainder); `data register / unregister` accept positional `<name>` and `<path>` alongside the `--name` / `--path` options, with conflict detection.
+- **`--log-level` plumbing complete** — `apply_logging_level` now sets the root logger so third-party libraries (transformers / peft / trl) actually respect QUIET and DEBUG. The four tiers no longer produce byte-identical output.
+- **+36 net new tests** across the new helpers, security-fix follow-ups (path-containment in `register_data`, `bench.py`, `infer.py --output`, symlink-reject in custom Space templates), and Windows cross-drive edge cases.
 
 ## Why Soup?
 

{soup_cli-0.40.0 → soup_cli-0.40.2}/README.md

@@ -9,6 +9,7 @@
 </p>
 
 <p align="center">
+  <a href="https://trysoup.dev">Website</a> &middot;
   <a href="#quick-start">Quick Start</a> &middot;
   <a href="#features">Features</a> &middot;
   <a href="#data-tools">Data Tools</a> &middot;
@@ -24,6 +25,7 @@
   <img src="https://img.shields.io/badge/license-Apache--2.0-blue" alt="Apache-2.0 License">
   <a href="https://github.com/MakazhanAlpamys/Soup/actions"><img src="https://img.shields.io/endpoint?url=https://gist.githubusercontent.com/MakazhanAlpamys/65fdc943f85f3b2c46ecddb415c2b779/raw/soup_tests.json" alt="Tests"></a>
   <a href="https://github.com/MakazhanAlpamys/Soup/actions"><img src="https://github.com/MakazhanAlpamys/Soup/actions/workflows/ci.yml/badge.svg" alt="CI"></a>
+  <a href="https://trysoup.dev"><img src="https://img.shields.io/badge/website-trysoup.dev-blue" alt="Website"></a>
 </p>
 
 ---
@@ -38,15 +40,20 @@ soup train
 
 ## What's New
 
-Latest highlights only. Full history: [GitHub Releases](https://github.com/MakazhanAlpamys/Soup/releases).
 
-**v0.40.0 — Preference Variety**: BCO trainer + a unified preference-loss surface so DPO / SimPO / ORPO / IPO / BCO live behind one config knob. Adds two opt-in DPO controls (β-schedule + ref-model regen) and a forward-looking multi-objective preference-loss surface.
+Latest highlights only. Full history: [GitHub Releases](https://github.com/MakazhanAlpamys/Soup/releases).
 
-- **BCO Trainer** — set `task: bco` for Binary Classifier Optimization. Same input format as DPO (`prompt + chosen + rejected`); rows are internally split to TRL's BCO unpaired schema. New `training.bco_beta` (default 0.1, gt=0). Template: `soup init --template bco`.
-- **Unified preference dispatcher** — set `task: preference` + `training.preference_loss: dpo|simpo|orpo|ipo|bco` to pick the loss without renaming your task. Legacy `task: dpo`, `task: simpo`, etc remain first-class — the new surface is additive, not a breaking collapse. Useful for hyperparameter sweeps over the loss type itself.
-- **KL-controlled DPO variants** — anneal β over training with `training.dpo_beta_schedule: linear|cosine|exponential` + `training.dpo_beta_end`. Periodically refresh the frozen reference model with the current student via `training.dpo_ref_regen_epochs: 2`. Both gated to DPO-family tasks (`dpo`, `ipo`, or `preference` with `preference_loss in {dpo, ipo}`); transformers backend only.
-- **Multi-objective preference loss** — define `training.preference_loss_weights: {dpo: 0.7, bco: 0.3}` to blend losses. 2–5 entries, weights must sum to 1. Schema-level surface ships now; live runtime weighted-loss combination deferred to v0.40.1 (`PreferenceTrainerWrapper.setup` raises `NotImplementedError` with a friendly message until then — same stub-then-live pattern as v0.27.0 MII / v0.37.0 multipack / v0.38.0 quant menu / v0.39.0 ReLoRA).
-- **Net +118 tests** (4538 → 4656) across BCO trainer + dispatcher + β schedule math + ref-model regen TOCTOU + multi-objective schema bounds.
+**v0.40.2 — Quick polish + carry-overs**: closes 3 originally-scheduled GitHub issues (#36 eval-gate dashboard row, #50 smarter `--hf-resume`, #51 custom HF Space templates) plus 7 v0.40.1 long-tail UX papercuts.
+
+- **`--hf-resume` prefers local newer** — `prepare_hf_resume` checks the highest local `checkpoint-N` against the highest remote branch and skips the snapshot download when the local copy is newer or equal. Saves bandwidth and never overwrites a fresher local checkpoint with stale Hub state.
+- **Eval-gate dashboard row** — pure formatter `format_gate_row(state)` lives in `soup_cli/monitoring/display.py` and renders a one-liner like `Gate: helpfulness 7.8 ✓ | math 0.82 ✗ (-0.06) | STOP` for the live training panel. Hidden when eval-gate is disabled.
+- **`soup deploy hf-space --template-dir <path>`** — supply your own `app.py` + `README.md` (+ optional `requirements.txt`) instead of the built-in `gradio-chat` / `streamlit-chat`. Containment-checked, repo-id substitution validated *before* render, 256 KB cap per file, symlinks rejected.
+- **`soup quickstart --output DIR`** — route data, config, and run dir under any directory you choose (default keeps backwards-compat in cwd).
+- **`soup runs --cwd-only`** — restrict the listing to runs whose `output_dir` is under the current directory; the global `~/.soup/experiments.db` view is still default.
+- **`soup infer` / `soup bench` accept HF ids** — when the local model path is missing AND the value isn't path-like (no `./`, `/`, `~`, `C:\`), it falls through to a HuggingFace download via `transformers.from_pretrained`. Path-like-but-missing surfaces a friendly `FileNotFoundError`.
+- **CLI flag aliases** — `data filter --min-coherence` (alias for `--coherence`); `data split --train` accepted (informational, train is the implicit remainder); `data register / unregister` accept positional `<name>` and `<path>` alongside the `--name` / `--path` options, with conflict detection.
+- **`--log-level` plumbing complete** — `apply_logging_level` now sets the root logger so third-party libraries (transformers / peft / trl) actually respect QUIET and DEBUG. The four tiers no longer produce byte-identical output.
+- **+36 net new tests** across the new helpers, security-fix follow-ups (path-containment in `register_data`, `bench.py`, `infer.py --output`, symlink-reject in custom Space templates), and Windows cross-drive edge cases.
 
 ## Why Soup?
 

{soup_cli-0.40.0 → soup_cli-0.40.2}/SECURITY.md

@@ -9,7 +9,8 @@ We provide security updates for the following versions:
 - **Versions older than 3 minor versions:** No support
 
 Example:
-- v0.40.0-0.40.x -- Full support (latest)
+- v0.40.2 -- Full support (latest)
+- v0.40.0-v0.40.x -- Full support
 - v0.39.0-0.39.x -- Bug-fix support only
 - v0.38.0-0.38.x -- Bug-fix support only
 - v0.37.x and below -- No support
@@ -143,6 +144,8 @@ No known critical vulnerabilities in current releases.
 - **v0.32.0 — Training Stability & Auto-Tuning**: `--find-lr-output` containment via shared `utils/paths.is_under_cwd` (prevents writes outside cwd); `save_lr_finder_report` rejects NaN / Infinity floats in `lrs` / `losses` and serialises with `allow_nan=False` (keeps the report parser-safe); `compute_lr_schedule` rejects non-positive `start_lr`, inverted ranges, and `num_steps` outside `[2, 10_000]`; `pick_mixed_precision` rejects empty / null-byte / >200-char model names and resolves multi-version quirks (`qwen2.5` vs `qwen2`, `phi-3.5` vs `phi-3`) by longest-substring-first iteration so an added family can never accidentally make a more-specific entry dead code; `compute_warmup_steps` clamps to `[10, 1000]` with a `ratio==0.0` short-circuit matching HF Trainer's "no warmup" convention; `SpikeRecoveryStrategy` is `@dataclass(frozen=True)` (post-construction mutation cannot bypass validation), `max_attempts ∈ [1, 10]`, `lr_decay ∈ (0, 1)`, `min_lr > 0`; cross-validator `_validate_spike_recovery_requires_watchdog` rejects `loss_spike_recovery=true, loss_watchdog=false` at config-load (fails fast instead of never triggering); `convergence_window ∈ [5, 10_000]`, `convergence_rel_tol ∈ (0, 1]`, `recommend_action` reuses `detect_plateau` so plateau heuristic stays single-source-of-truth; `GradAccumMonitor.recommend()` caps doubled `accum` at `MAX_ACCUM=1024` so a runaway advisory loop cannot blow up DataLoader prefetch; `generate_config` validates BOTH the YAML output path AND the embedded `decisions["output"]` field via `is_under_cwd` (closes the gap where a crafted `decisions["output"]="../../etc"` would have silently propagated into the rendered YAML)
 - **v0.34.0 — Observability & Dev UX**: `.crash` bundle generator (`utils/crash.py`) recursively redacts `hf_*` / `sk-*` / `Bearer …` token-shaped strings in any captured `config` and metric tail before serialisation, so a `.crash` file shared on a public GitHub issue cannot leak credentials; `output_dir` is reduced to `os.path.basename` so `$HOME` doesn't leak; `write_crash_bundle` uses `os.path.realpath + commonpath` for cwd containment (Windows-safe; raises `ValueError` not `PermissionError` so callers cannot silently swallow with `except OSError`); filename appends `secrets.token_hex(4)` so two crashes in the same UTC second don't collide; bundle truncated to `MAX_BUNDLE_BYTES=1_000_000`. `train.py` crash-write surfaces failures to the user (no silent missing-bundle). `profiling.py` `resolve_trace_path` rejects empty / `.` / `..` / `/` / `\\` / null-byte `run_id` (closes the `output_dir/profiles/../trace.json` escape) and uses `os.path.realpath + is_under_cwd`; profiles dir is created only on successful torch import (no stale empty dirs on torch-less CI). `tracker.get_run` LIKE-prefix match escapes `%` / `_` / `\\` and uses `ESCAPE '\\'` so a crafted `run_id` cannot widen the match (mirrors v0.26.0 registry policy). Lazy schema migration (`_ensure_schema`) tolerates the "duplicate column" race when two CLI processes start simultaneously on a fresh DB (fork-based multi-GPU training, TUI auto-refresh). `runs.py show/replay/clean` switched user `run_id` rendering to `markup_escape` and switched `clean` containment from broken `Path.resolve() + relative_to()` to project-standard `os.path.realpath + is_under_cwd`. `tui_app.py` lazy-imports `ExperimentTracker` and `markup_escape`s every DB-sourced string before passing into Textual widgets so a crafted base_model / experiment_name cannot inject `[bold red]…[/]` markup. `run_cost.estimate_run_cost_usd` rejects `bool` in `num_gpus` (bool is a subclass of int — same defence as v0.30.0 `Candidate.__post_init__`); duration clamped to `[0, 1 year]`; unknown GPU returns `None` so callers render `—` instead of fabricating `$0.00`. `log_level.parse_log_level` rejects non-string + null-byte input.
 - **v0.33.0 — Live Wire**: RLVR `code_exec_reward` adds OS-level isolation (Linux best-effort `os.unshare(CLONE_NEWUSER|CLONE_NEWNET|CLONE_NEWPID)`, macOS `sandbox-exec` with default-deny `MACOS_SANDBOX_PROFILE` narrowed to a 3-name `mach-lookup` allowlist to prevent DNS / NSURLSession bypass of `(deny network*)`); `prune_checkpoints` switches to TOCTOU-safe `os.lstat + S_ISLNK` + `shutil.rmtree(onerror=_abort_on_symlink)` so a symlink encountered mid-walk aborts rather than escapes; `run_gate` wraps each task scorer in a typed `try/except` so backend failures produce `score=None, error=str(exc)` (never silent `score=1.0`); `_parse_judge_url` removes the bare `http://` catch-all (defence-in-depth after the Pydantic GateTask validator); `soup can run` requires `--yes` or explicit consent callback and raises `ValueError` (not `PermissionError`, which is an `OSError` subclass that broad `except` blocks would swallow); GGUF `rglob` result for ollama deploy is `realpath+commonpath` checked against extract_dir (prevents symlink escape from a crafted can); `DeployTarget.path` validator normalises mixed `\\`/`/` separators before splitting (closes a Windows `..` bypass); `CAN_FORMAT_VERSION` 1→2 (additive — v1 still loads); `soup can publish` validates `repo_id` via `utils/hf.validate_repo_id`, resolves token via `resolve_token`, sanitises commit messages (first-line, 200-char cap), uses HTTPS-only HfApi; `_write_spike_recovery_hint` adds `is_under_cwd` containment check on `args.output_dir` from raw HF `TrainingArguments`; `lookup_entry_by_output_dir` emits `ResourceWarning` when 1000-row scan limit is hit (no silent miss); `CrossDocCollator` no longer mutates input feature dicts (HF Dataset rows are cached and reused — mutation broke subsequent batches); `Candidate` rejects `bool` in `score`/`latency_ms` (was sneaking past `int` isinstance check); `evaluate_candidate` latency mean now divides by *completed* prompts (excludes crashed) so a broken candidate isn't artificially fast; `auto_quant.run_auto_quant_picker` soft-falls-back to highest-scored candidate when no candidate clears `min_score` (server still binds); `build_logits_processors` returns `[]` when neither `outlines` nor `lm-format-enforcer` is installed (server degrades to free-form rather than 500); MII server uses loopback-only CORS, max_tokens cap [1, 16384], stream rejection, generic 500 with no stack-trace leak; `os.execvp` auto-reexec uses list args (no shell), all forwarded flags pre-validated; `cleanup_extract_dir` uses `os.path.commonpath` (Windows-safe) instead of `startswith`; `_run_subprocess` catches `TimeoutExpired` and returns rc=124 (coreutils convention) instead of an unhandled traceback; new `eval_results` and `tensorrt` artifact kinds in `RegistryStore._VALID_KINDS`
+- **v0.40.2 — Quick polish + carry-overs**: New `soup_cli/utils/hf_space.py:render_custom_template_dir` enforces `is_under_cwd` containment on the template directory; `validate_repo_id` runs BEFORE `{MODEL_REPO}` substitution (matches v0.29.0 Part F policy); per-file 256 KB cap (matches v0.39.0 Part E template-size policy); only `app.py` / `README.md` / `requirements.txt` are read (closed allowlist — no path-from-user-data). Symlinks rejected via `os.lstat + stat.S_ISLNK` and non-regular files (FIFO / device) also rejected (matches v0.33.0 #22 prune_checkpoints TOCTOU policy) — defends against `<template_dir>/app.py -> /etc/passwd`. `_find_highest_local_checkpoint` reads `output_dir` after caller's `is_under_cwd` validation (in `prepare_hf_resume`) and silently drops non-directories + OSError. `prepare_hf_resume` skips the snapshot download when local `checkpoint-N >= remote checkpoint-N` (saves bandwidth and never overwrites a fresher local checkpoint). `commands/data.py:register_data` containment switched from `Path.resolve() + relative_to()` to shared `is_under_cwd` (Windows 8.3 short-name safety per CLAUDE.md project rule); same fix applied to `commands/bench.py` prompts-file containment. `commands/infer.py:--output` now containment-checked via `is_under_cwd` (late-evaluated after model+input validation so pre-existing `tmp_path` test contracts keep working). `commands/quickstart.py:--output` validates target dir via `is_under_cwd` before `mkdir(parents=True)`; rejects out-of-cwd targets with friendly message. `commands/runs.py:_filter_runs_by_cwd` uses `os.path.realpath + commonpath`, catches `(ValueError, OSError)` so cross-drive paths on Windows (`D:\runs` vs `C:\project`) drop silently rather than crash. `monitoring/display.py:format_gate_row` uses explicit `task.get("passed") is True` so a missing `"passed"` field renders neutrally instead of as a false-y red ✗. `commands/infer.py:_resolve_model_source` heuristic for HF-id-vs-local-path: only falls through to HF when value is NOT path-like (no `./`, `/`, `\\`, `~`, no Windows drive letter, non-empty); path-like-but-missing raises `FileNotFoundError` so users see actionable errors instead of confusing HF download attempts. Known limitation: custom HF Space templates always create the Space with `space_sdk="gradio"` regardless of the supplied `app.py` (no `--sdk` flag in this release; combine `--template streamlit-chat` with the inline registry for Streamlit Spaces). Tracked for v0.40.3+.
+- **v0.40.1 — QA Hardening**: `soup_cli/utils/encoding.force_utf8_stdio` reconfigures Windows stdout/stderr to UTF-8 before any Rich Console is constructed; `os.environ.setdefault("PYTHONIOENCODING", "utf-8")` preserves user override; `(OSError, ValueError, AttributeError)` swallowed on detached streams; POSIX no-op. `SoupConfig._remap_root_level_misplaced_keys` (model_validator, mode='before') migrates root-level `lora:` into `training.lora` so nested validators (including `lora.init_strategy: Literal["random","pissa","olora"]`) actually fire — closes a footgun where the misplaced key was silently dropped. Caller's dict is never mutated (shallow-copy policy mirroring v0.33.0 #47 / v0.40.0 Part B). `PreferenceTrainerWrapper._build_multi_objective` replaces the v0.40.0 `NotImplementedError` stub with a primary-loss approximation; `validate_weight_compat` rejects BCO mixed with paired losses at runtime (data-format incompatible). `combine_losses` rejects empty weights, propagates NaN loudly (no silent zeroing), and rejects `bool` weight values (matches v0.30.0 `Candidate` policy). `_probe_cache_param_count` rejects empty / null-byte model names before path construction (mirrors v0.26.0 registry / v0.39.0 ReLoRAPolicy policy). `commands/doctor` flags `transformers ≥ 5.0.0` as INCOMPATIBLE via `_MAX_EXCLUSIVE` table; `_version_ge` parses leading-int chunks so `5.0.0.dev0` correctly trips the cap. `_detect_gpu_hw_without_torch_cuda` calls `nvidia-smi` via argv list (no shell), 5s timeout, `OSError` / `TimeoutExpired` caught; GPU label from `nvidia-smi` stdout is `rich.markup.escape`d before embedding in Rich-markup string (a real GPU name like `NVIDIA Quadro [T4]` cannot break or inject markup). `_detect_dual_python_interpreters` uses `os.path.realpath` (not `Path.resolve()`) for Windows 8.3 short-name compat. `_pick_quickstart_model` swaps TinyLlama-1.1B → SmolLM2-135M when `total_memory ≤ 6 GB` (prevents step-0 OOM on RTX 3050 4 GB / similar). `_live_lr_sweep_from_config` switched broken `load_local` import to `load_raw_data` (previously always silently fell back to a static placeholder curve). `commands/migrate` rejects `.jsonl` input (with first-line `{` sniff) with exit-2 friendly error; `.jsonl`-only suffix gate prevents false-positives on `.ipynb` notebooks. `commands/eval custom -o` is now honored independently of `--attach-to-registry`; loop-shadow regression where `output = generate_fn(...)` overwrote the CLI option fixed (variable renamed to `response`). `_load_jsonl` switched from `utf-8` to `utf-8-sig` so PowerShell `Out-File -Encoding utf8`-produced JSONL no longer fails first-row parse. Known limitation: `--trust-remote-code` opt-in surface still excludes 10 non-SFT trainers + 5 commands (v0.36.0 #63 carry-over).
 - **v0.40.0 — Preference Variety**: New `task='bco'` (Binary Classifier Optimization) and `task='preference'` (unified dispatcher). New schema fields: `bco_beta` (gt=0), `preference_loss: Literal[dpo,simpo,orpo,ipo,bco]|None`, `preference_loss_weights: Optional[Dict[str,float]]`, `dpo_beta_schedule: Literal[linear,cosine,exponential]|None`, `dpo_beta_end: float, gt=0|None`, `dpo_ref_regen_epochs: int [1,1000]|None`. Cross-validators: `_validate_preference_dispatcher` rejects setting either `preference_loss` or `preference_loss_weights` outside `task='preference'` (closes ordering-dependency between Part B/D validators); `_validate_dpo_variants_supported_tasks` gates β-schedule + ref-regen to DPO-family tasks (`dpo`, `ipo`, or `preference` + `preference_loss in {dpo, ipo}`); rejected on mlx backend with distinct error message (matches v0.34.0 distinct-reason policy); `_validate_preference_loss_weights` enforces 2–5 entries (single-entry rejected with actionable message pointing at scalar `preference_loss`), key allowlist `{dpo, simpo, orpo, ipo, bco}`, explicit null-byte rejection on keys (matches v0.39.0 rank_pattern policy), per-value bounds `(0, 1]`, weights must sum to 1.0 (±1e-6), mutually exclusive with scalar `preference_loss`, rejected on mlx backend. `compute_beta_at_step` rejects `bool` on `step` and `total_steps` (project bool-as-int policy from v0.30.0). `BetaScheduleCallback` resolves `total_steps` lazily in `on_train_begin` so the schedule sees the real `state.max_steps` populated by HF Trainer (closes a first-cut silent-no-op bug where total_steps=0 emitted beta_end for every step). `RefModelRegenCallback._regenerate` uses `strict=True` on `load_state_dict` and logs at WARNING on mismatch (closes a first-cut silent partial-copy hazard where strict=False could produce a hybrid old-base + new-LoRA reference); epoch 0 regen suppressed (avoids copying untrained student); trainer `.beta` assignment swallow narrowed to `AttributeError` only. `PreferenceTrainerWrapper._make_inner_cfg` uses `model_copy` (not `model_dump`+`model_validate`) so re-validation never sees an inconsistent intermediate state and the caller's `cfg` is never mutated (mirrors v0.33.0 #47 immutability policy). `_split_dpo_rows_to_bco` skipped-row count emitted at DEBUG so production silent-degradation is inspectable (mirrors v0.33.0 #47 CrossDocCollator policy). Multi-objective live runtime weighted-loss combination is deferred to v0.40.1: `PreferenceTrainerWrapper.setup` raises `NotImplementedError` with a friendly message naming the deferred-version follow-up (mirrors v0.27.0 MII / v0.37.0 multipack / v0.38.0 quant menu / v0.39.0 ReLoRA stub-then-live pattern). Known limitation: `BCOTrainerWrapper._setup_transformers` still hardcodes `trust_remote_code=True` (v0.36.0 #63 known-gap family carry-over across non-SFT trainers).
 - **v0.39.0 — LoRA Quality**: `LoraConfig.init_strategy: Literal["random","pissa","olora"]` rejects unknown strategies; PiSSA + DoRA / VeRA combinations rejected at config-load. `model_validator(mode="before")` aligns `use_olora=True` → `init_strategy="olora"` via dict-copy (no caller mutation; matches v0.33.0 #47 immutability policy). `rank_pattern`/`alpha_pattern: Optional[Dict[str, int]]` capped at 256 keys × value (0, 1024], rejects `bool` (subclass of `int` — matches v0.30.0 `Candidate` policy), null bytes in keys, empty keys; cross-validator rejects with `use_vera=True`. `ReLoRAPolicy` is `@dataclass(frozen=True)` (post-construction mutation raises `FrozenInstanceError`); bounds: `steps ∈ [1, 1e7]`, `warmup_ratio ∈ [0, 1]`, `prune_ratio ∈ (0, 1)` (strict — prevents zero-everything footgun). `magnitude_prune_tensor` strict `0 < prune_ratio < 1` rejection, non-Tensor input raises `TypeError`, empty / single-element tensor short-circuits (avoids `kthvalue(_, 0)` runtime crash). `_validate_relora_supported_tasks` cross-validator rejects `relora_steps` with `task != "sft"` and `backend=mlx` with distinct error messages (matches v0.34.0 distinct-reason policy); multi-trainer expansion deferred to v0.39.1. `is_gemma4_model` uses a word-boundary regex (`(?:^|[^a-z0-9])gemma-?4(?:[^a-z0-9]|$)`) so `"ungemma4ed"` / `"my-gemma4ish"` no longer over-match; null-byte rejection on `model_name`. `apply_gemma4_clippable_patch` weight-copy fallback logs at DEBUG instead of silent random-init; the patch is gated by `is_gemma4_model(cfg.base)` in `sft.py` before invocation so non-Gemma4 trainings never traverse the module tree. `apply_surgical_patches` rejects empty / null-byte `model_name` with `ValueError`. `templates/load_template` containment: filename re-validated via `_validate_name` (rejects `..`/`/`/`\\`/null/empty); `os.path.realpath + os.path.commonpath` containment check on the resolved path against `_templates_dir()` so a tampered `manifest.json` cannot read files outside the package directory (mirrors v0.26.0 registry policy). Tampered-manifest `ValueError` from `_validate_name` caught and falls back to inline (no propagating exception). 256 KB file-size cap. Inline `TEMPLATES` carries an explicit deprecation comment pointing at the canonical YAML registry; `tests/test_templates_yaml.py` asserts byte-equality of all 16 inline ↔ YAML pairs to prevent silent drift. Planned removal: v0.41.0+.
 - **v0.38.0 — Quant Menu**: `TrainingConfig.quantization` Literal extended with `gptq` / `awq` / `hqq:1bit`..`hqq:8bit` (no `hqq:7bit` — HQQ doesn't support it) / `aqlm` / `eetq` / `mxfp4` / `fp8`; Pydantic rejects every other string at config-load. `validate_gptq_checkpoint` and `validate_awq_checkpoint` probe local paths for `quantize_config.json` / `quant_config.json`; HF repo IDs fall through; null-byte rejection + non-string `TypeError` on the ref. `_validate_prequantized_no_qat` rejects every pre-quantized format combined with `quantization_aware` (int8 QAT or `'fp8'`) — pre-quantized weights carry their own scale and QAT/FP8 prepare would silently corrupt them (mirrors LlamaFactory `quantization.py:117/199/211`). `_validate_bnb_quant_storage_only_with_4bit` rejects `bnb_4bit_quant_storage` on every non-BNB-4bit format (silent no-op otherwise); allowed dtypes: `Literal["uint8", "float16", "bfloat16", "float32"]`. `_validate_quant_menu_supported_tasks` restricts the new formats to `task='sft'` on `backend='transformers'` in v0.38.0 with distinct MLX-backend vs unsupported-task error messages (matches v0.34.0 distinct-reason policy). `check_quant_distributed_compat` hard-fails HQQ/EETQ/AQLM × {FSDP, ZeRO-3} (sourced from LlamaFactory `quantization.py:199/211` plus AQLM dequant constraints); warning-tier (not error) for BNB-4bit + FSDP without `bnb_4bit_quant_storage` so users see the silent perf cliff; unknown `quantization` raises `ValueError` (no silent pass) and the check is wired into `commands/train.py` startup. `parse_hqq_bits` rejects unsupported bit-rates and malformed `hqq:` strings before any kernel build.

{soup_cli-0.40.0 → soup_cli-0.40.2}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "soup-cli"
-version = "0.40.0"
+version = "0.40.2"
 description = "Fine-tune LLMs in one command. No SSH, no config hell."
 readme = "README.md"
 license = "Apache-2.0"
@@ -27,7 +27,7 @@ dependencies = [
     "pydantic>=2.0.0",
     "pyyaml>=6.0",
     "torch>=2.0.0",
-    "transformers>=4.36.0",
+    "transformers>=4.36.0,<5.0.0",
     "peft>=0.7.0",
     "trl>=0.7.0",
     "datasets>=2.14.0",

{soup_cli-0.40.0 → soup_cli-0.40.2}/soup_cli/__init__.py

@@ -1,3 +1,3 @@
 """Soup CLI — Fine-tune LLMs in one command."""
 
-__version__ = "0.40.0"
+__version__ = "0.40.2"

{soup_cli-0.40.0 → soup_cli-0.40.2}/soup_cli/autopilot/analyzer.py

@@ -91,12 +91,65 @@ _MODEL_SIZE_RE = re.compile(r"(\d+(?:\.\d+)?)\s*[Bb]")
 
 
 def _guess_params_from_name(name: str) -> float:
-    """Extract the parameter count (in billions) from a model name."""
+    """Extract the parameter count (in billions) from a model name.
+
+    v0.40.1 Part C / C3 — fall back to **1B** (not 7B) when the name has no
+    embedded size hint. The previous 7.0 default made tiny models like
+    ``tiny-gpt2`` (5 MB) fail VRAM-budget checks on machines that could
+    trivially train them. We also recognise the ``-Mm`` (millions) format
+    used by SmolLM2 / Phi-3 family. Probes for a local safetensors index
+    (cached HF snapshot) before falling back.
+    """
     match = _MODEL_SIZE_RE.search(name)
     if match:
         return float(match.group(1))
-    # Conservative default
-    return 7.0
+    # Recognise <N>m / <N>M for sub-billion models (e.g. SmolLM2-135M,
+    # tiny-gpt2). Convert to billions.
+    m_match = re.search(r"(\d+(?:\.\d+)?)\s*[Mm](?![a-zA-Z])", name)
+    if m_match:
+        return float(m_match.group(1)) / 1000.0
+    # Probe local HF cache for safetensors index size if we can.
+    cache_size = _probe_cache_param_count(name)
+    if cache_size is not None:
+        return cache_size
+    # Conservative default — small assumption (yellow advisory should fire).
+    return 1.0
+
+
+def _probe_cache_param_count(name: str) -> Optional[float]:
+    """Best-effort: read parameter count from cached safetensors index.
+
+    Looks at ``~/.cache/huggingface/hub/models--<owner>--<repo>/snapshots/*/
+    model.safetensors.index.json`` and returns ``total_size / 4 / 1e9`` (fp32
+    bytes per param). Returns ``None`` if not found.
+
+    v0.40.1 review fix — reject empty / null-byte names (project policy
+    mirroring v0.26.0 registry / v0.39.0 ReLoRAPolicy) before constructing
+    the cache path.
+    """
+    if not isinstance(name, str) or not name or "\x00" in name:
+        return None
+    try:
+        from pathlib import Path as _Path
+
+        owner_repo = name.replace("/", "--")
+        cache = _Path.home() / ".cache" / "huggingface" / "hub" / f"models--{owner_repo}"
+        if not cache.is_dir():
+            return None
+        for idx in cache.rglob("model.safetensors.index.json"):
+            try:
+                import json as _json
+                data = _json.loads(idx.read_text(encoding="utf-8"))
+                total_bytes = data.get("metadata", {}).get("total_size")
+                if isinstance(total_bytes, (int, float)) and total_bytes > 0:
+                    # Assume fp32 storage (4 bytes/param) — generous upper
+                    # bound; bf16/fp16 cuts it in half.
+                    return float(total_bytes) / 4.0 / 1e9
+            except (OSError, ValueError):
+                continue
+    except Exception:  # noqa: BLE001
+        return None
+    return None
 
 
 def analyze_model(name: str, params_b: Optional[float] = None) -> ModelProfile:

{soup_cli-0.40.0 → soup_cli-0.40.2}/soup_cli/cli.py

@@ -2,11 +2,20 @@
 
 import sys
 
-import typer
-from rich.console import Console
+# UTF-8 stdio bootstrap (v0.40.1 Part A) — must run before any Rich console
+# is constructed. On Windows, reconfigures sys.stdout/stderr to UTF-8 so β /
+# ✓ / box-drawing chars don't crash with UnicodeEncodeError on cp1251/cp1252.
+# POSIX: no-op.
+from soup_cli.utils.encoding import force_utf8_stdio
 
-from soup_cli import __version__
-from soup_cli.commands import (
+force_utf8_stdio()
+_utf8_bootstrap_done = True
+
+import typer  # noqa: E402
+from rich.console import Console  # noqa: E402
+
+from soup_cli import __version__  # noqa: E402
+from soup_cli.commands import (  # noqa: E402
     adapters,
     autopilot,
     bench,
@@ -34,15 +43,15 @@ from soup_cli.commands import (
     train,
     ui,
 )
-from soup_cli.commands import doctor as doctor_cmd
-from soup_cli.commands import quickstart as quickstart_cmd
-from soup_cli.commands import (
+from soup_cli.commands import doctor as doctor_cmd  # noqa: E402
+from soup_cli.commands import quickstart as quickstart_cmd  # noqa: E402
+from soup_cli.commands import (  # noqa: E402
     tui as tui_cmd,
 )
-from soup_cli.commands import (
+from soup_cli.commands import (  # noqa: E402
     why as why_cmd,
 )
-from soup_cli.utils.constants import GITHUB_URL
+from soup_cli.utils.constants import GITHUB_URL  # noqa: E402
 
 console = Console()
 
@@ -217,7 +226,11 @@ def main(
     """Soup — fine-tune LLMs in one command."""
     global _verbose, _log_level
     _verbose = verbose
-    from soup_cli.utils.log_level import parse_log_level, setup_logging
+    from soup_cli.utils.log_level import (
+        apply_logging_level,
+        parse_log_level,
+        setup_logging,
+    )
 
     try:
         tier = parse_log_level(log_level)
@@ -226,6 +239,10 @@ def main(
         raise typer.Exit(code=2) from exc
     _log_level = tier.value
     setup_logging(tier)
+    # v0.40.2 N1/G2: also push the tier into the root logger so third-party
+    # libraries (transformers / peft / trl) respect QUIET / DEBUG. Without
+    # this, all four levels were producing nearly-identical output.
+    apply_logging_level(tier)
 
 
 def run():

{soup_cli-0.40.0 → soup_cli-0.40.2}/soup_cli/commands/bench.py

@@ -43,13 +43,24 @@ def bench(
     """Run an inference benchmark (speed and memory) on a loaded model."""
     import torch
 
-    from soup_cli.commands.infer import _generate, _load_model
+    from soup_cli.commands.infer import _generate, _load_model, _resolve_model_source
     from soup_cli.utils.gpu import detect_device
 
-    model_path = Path(model)
-    if not model_path.exists():
-        console.print(f"[red]Model not found: {model_path}[/]")
-        raise typer.Exit(1)
+    # Resolve local-path-or-HF-id (#N7).
+    try:
+        model_kind, model_ref = _resolve_model_source(model)
+    except FileNotFoundError as exc:
+        console.print(
+            f"[red]{exc}[/]\n"
+            "[dim]If you meant a HuggingFace repo, use the form "
+            "'owner/repo-name' (no leading './').[/]"
+        )
+        raise typer.Exit(1) from exc
+    model_path = Path(model_ref)
+    if model_kind == "hf":
+        console.print(
+            f"[dim]Local path not found; treating {model_ref!r} as a HF repo id.[/]"
+        )
 
     device, _ = detect_device()
 
@@ -61,15 +72,17 @@ def bench(
 
     if prompts_file:
         import json
-        p_path = Path(prompts_file).resolve()
+        import os as _os
 
-        try:
-            p_path.relative_to(Path.cwd())
-        except ValueError:
+        from soup_cli.utils.paths import is_under_cwd
+
+        if not is_under_cwd(prompts_file):
             console.print(
-                f"[red]Security Error:[/] Path {p_path} is outside the current working directory."
+                "[red]Security Error:[/] Prompts file must stay under the "
+                "current working directory."
             )
             raise typer.Exit(1)
+        p_path = Path(_os.path.realpath(prompts_file))
 
         if not p_path.is_file():
             console.print(f"[red]Prompts file not found:[/] {p_path}")

{soup_cli-0.40.0 → soup_cli-0.40.2}/soup_cli/commands/data.py

@@ -5,6 +5,7 @@ from __future__ import annotations
 import json
 import random
 from pathlib import Path
+from typing import Optional
 
 import typer
 from rich.console import Console
@@ -325,7 +326,7 @@ def filter_data(
         help="Max perplexity threshold (rows above this are removed)",
     ),
     coherence: float = typer.Option(
-        None, "--coherence",
+        None, "--coherence", "--min-coherence",
         help="Min coherence threshold 0.0-1.0 (rows below this are removed)",
     ),
     perplexity_model: str = typer.Option(
@@ -746,8 +747,11 @@ def sample_data(
         sampled = _sample_random(data, sample_count, seed=seed)
 
     # Resolve output path (with path traversal protection on explicit --output)
+    # v0.40.1 Part E — include the strategy in the default filename so
+    # successive `random` / `diverse` / `hard` runs don't silently overwrite
+    # each other.
     if output is None:
-        out_path = file_path.parent / f"{file_path.stem}_sampled.jsonl"
+        out_path = file_path.parent / f"{file_path.stem}_sampled_{strategy}.jsonl"
     else:
         out_path = Path(output).resolve()
         cwd = Path.cwd().resolve()
@@ -777,6 +781,13 @@ def split_data(
         None, "--test",
         help="Test split: percentage (default) or absolute count (with --absolute)",
     ),
+    train: int = typer.Option(
+        None, "--train",
+        help=(
+            "Train split (informational; the train remainder is implied by "
+            "--val + --test). Accepted for command parity."
+        ),
+    ),
     absolute: bool = typer.Option(
         False, "--absolute",
         help="Treat --val/--test as absolute sample counts instead of percentages",
@@ -1458,56 +1469,104 @@ def _load_augment_provider(provider: str, rpm: int):
 
 @app.command(name="register")
 def register_data(
-    name: str = typer.Option(..., "--name", "-n", help="Dataset name"),
-    path: str = typer.Option(..., "--path", "-p", help="Path to dataset file"),
+    name_arg: Optional[str] = typer.Argument(
+        None, help="Dataset name (positional alternative to --name)",
+    ),
+    path_arg: Optional[str] = typer.Argument(
+        None, help="Path to dataset file (positional alternative to --path)",
+    ),
+    name: Optional[str] = typer.Option(None, "--name", "-n", help="Dataset name"),
+    path: Optional[str] = typer.Option(
+        None, "--path", "-p", help="Path to dataset file"
+    ),
     fmt: str = typer.Option(
         "auto", "--format", "-f",
         help="Dataset format: alpaca, sharegpt, chatml, dpo, kto, auto",
     ),
 ):
-    """Register a local dataset by name for use in soup.yaml."""
+    """Register a local dataset by name for use in soup.yaml.
+
+    Accepts both ``--name X --path Y`` and positional ``X Y``.
+    """
     from soup_cli.utils.registry import register_dataset
 
-    # Path traversal protection
-    resolved = Path(path).resolve()
-    cwd = Path.cwd().resolve()
-    try:
-        resolved.relative_to(cwd)
-    except ValueError:
+    # Resolve positional vs option, with conflict detection
+    final_name = name if name is not None else name_arg
+    final_path = path if path is not None else path_arg
+    if final_name is None or final_path is None:
+        console.print(
+            "[red]Provide both name and path "
+            "(positional `<name> <path>` or `--name --path`).[/]"
+        )
+        raise typer.Exit(2)
+    if name is not None and name_arg is not None and name != name_arg:
+        console.print("[red]Conflict: --name and positional name differ.[/]")
+        raise typer.Exit(2)
+    if path is not None and path_arg is not None and path != path_arg:
+        console.print("[red]Conflict: --path and positional path differ.[/]")
+        raise typer.Exit(2)
+
+    # Path traversal protection — use os.path.realpath + commonpath
+    # (project standard; Windows 8.3 short-name safe).
+    import os as _os
+
+    from soup_cli.utils.paths import is_under_cwd
+
+    if not is_under_cwd(final_path):
         console.print(
             "[red]Dataset path must be under the current working directory.[/]"
         )
         raise typer.Exit(1)
+    resolved = Path(_os.path.realpath(final_path))
 
     registry_path = _get_registry_path()
 
     try:
-        register_dataset(name, str(resolved), fmt, registry_path=registry_path)
+        register_dataset(final_name, str(resolved), fmt, registry_path=registry_path)
     except ValueError as exc:
         console.print(f"[red]{exc}[/]")
         raise typer.Exit(1)
 
     console.print(
-        f"[green]Registered dataset '[bold]{name}[/bold]'[/]\n"
-        f"  Path: {path}\n"
+        f"[green]Registered dataset '[bold]{final_name}[/bold]'[/]\n"
+        f"  Path: {final_path}\n"
         f"  Format: {fmt}"
     )
 
 
 @app.command(name="unregister")
 def unregister_data(
-    name: str = typer.Option(..., "--name", "-n", help="Dataset name to remove"),
+    name_arg: Optional[str] = typer.Argument(
+        None, help="Dataset name (positional alternative to --name)",
+    ),
+    name: Optional[str] = typer.Option(
+        None, "--name", "-n", help="Dataset name to remove"
+    ),
 ):
-    """Remove a dataset from the local registry."""
+    """Remove a dataset from the local registry.
+
+    Accepts both ``--name X`` and positional ``X``.
+    """
     from soup_cli.utils.registry import unregister_dataset
 
+    final_name = name if name is not None else name_arg
+    if final_name is None:
+        console.print(
+            "[red]Provide a dataset name "
+            "(positional `<name>` or `--name`).[/]"
+        )
+        raise typer.Exit(2)
+    if name is not None and name_arg is not None and name != name_arg:
+        console.print("[red]Conflict: --name and positional name differ.[/]")
+        raise typer.Exit(2)
+
     registry_path = _get_registry_path()
-    removed = unregister_dataset(name, registry_path=registry_path)
+    removed = unregister_dataset(final_name, registry_path=registry_path)
 
     if removed:
-        console.print(f"[green]Removed dataset '{name}' from registry.[/]")
+        console.print(f"[green]Removed dataset '{final_name}' from registry.[/]")
     else:
-        console.print(f"[red]Dataset '{name}' not found in registry.[/]")
+        console.print(f"[red]Dataset '{final_name}' not found in registry.[/]")
         raise typer.Exit(1)
 
 

{soup_cli-0.40.0 → soup_cli-0.40.2}/soup_cli/commands/deploy.py

@@ -435,6 +435,15 @@ def hf_space(
         "-t",
         help=f"Space template: {', '.join(HF_SPACE_TEMPLATES.keys())}",
     ),
+    template_dir: Optional[str] = typer.Option(
+        None,
+        "--template-dir",
+        help=(
+            "Custom template directory (overrides --template). "
+            "Must contain app.py + README.md, optionally requirements.txt. "
+            "Use {MODEL_REPO} placeholder for substitution."
+        ),
+    ),
     private: bool = typer.Option(
         False, "--private", help="Create the Space as private",
     ),
@@ -461,7 +470,7 @@ def hf_space(
     except ValueError as exc:
         console.print(f"[red]Invalid --space repo id:[/] {exc}")
         raise typer.Exit(1) from exc
-    if template not in HF_SPACE_TEMPLATES:
+    if template_dir is None and template not in HF_SPACE_TEMPLATES:
         console.print(
             f"[red]Unknown template: {template}[/]\n"
             f"Available: {', '.join(HF_SPACE_TEMPLATES.keys())}"
@@ -485,16 +494,28 @@ def hf_space(
 
     # --- Render template files ---
     try:
-        files = render_space_template(template, model_repo=model)
+        if template_dir is not None:
+            from soup_cli.utils.hf_space import render_custom_template_dir
+            files = render_custom_template_dir(template_dir, model_repo=model)
+            # Custom templates default to gradio SDK unless requirements
+            # imply otherwise; we record gradio for create_repo space_sdk.
+            sdk = "gradio"
+        else:
+            files = render_space_template(template, model_repo=model)
+            sdk = HF_SPACE_TEMPLATES[template]["sdk"]
     except ValueError as exc:
         console.print(f"[red]Template render failed:[/] {exc}")
         raise typer.Exit(1) from exc
+    except FileNotFoundError as exc:
+        console.print(f"[red]Template directory error:[/] {exc}")
+        raise typer.Exit(1) from exc
 
+    template_label = template_dir if template_dir is not None else template
     console.print(
         Panel(
             f"Space:    [bold]{space}[/]\n"
             f"Model:    [bold]{model}[/]\n"
-            f"Template: [bold]{template}[/]\n"
+            f"Template: [bold]{template_label}[/]\n"
             f"Private:  [bold]{private}[/]",
             title="Deploy HuggingFace Space",
         )
@@ -514,7 +535,7 @@ def hf_space(
     try:
         api.create_repo(
             repo_id=space, repo_type="space",
-            space_sdk=HF_SPACE_TEMPLATES[template]["sdk"],
+            space_sdk=sdk,
             private=private, exist_ok=True,
         )
         for in_repo_name, content in files.items():