v0.8.0 — Universal pull: evaluate & serve any safetensors HF model

The headline: the stack no longer only serves a fixed curated list. Point it at any safetensors Hugging Face repo and it tells you — honestly, before you download anything — whether it'll run on your GPUs, and if so boots a working server for it.

What you can do now

# Will this model fit on my GPUs? (no download, no boot — just the verdict)
scripts/pull.sh <org/Model> --profile-like vllm/minimal --dry-run

# If it passes: download + generate a minimal compose + boot it
scripts/pull.sh <org/Model> --profile-like vllm/minimal --yes

• Any safetensors model, not just the catalog — if it passes the gates you get a working compose + a booted server, no manual compose authoring.
• Answered before you download — the fit verdict comes from precise KV math on the repo's own config, so you don't pull 20–50 GB to find out it won't fit.
• Honest, never silently wrong — every answer carries a confidence tier and a boot-fit ≠ runtime-stability caveat (a "fits" verdict is a boot-time check; validate with soak-continuous before trusting it for long agent workloads). Non-fits stop with a precise reason, not a cryptic crash.
• It improves over time — failures are classified and deduped into a calibration backbone, so the math sharpens without waiting for a per-model release.

Full user guide → docs/PULL.md (Quickstart at the top).

Honest limits (read before relying on it)

• safetensors + vLLM only. GGUF / .bin repos hard-block with a clear message (not a crash) — GGUF→llama.cpp is not served via pull (deferred).
• Trust is maintainer-promoted this phase — consensus automation is later.
• The curated path is unchanged — existing SINGLE_CARD / DUAL_CARD / catalog composes work exactly as before; pull is purely additive.

Commit-level detail below (auto-generated):

v0.8.0 — 2026-05-17

⚠️ Cliffs, gotchas, regressions

• v0.8.0 Pull-Gate P4-fix: price Tier-1 curated via curated-exact kv-calc spec, not generic-dense (+ non-mocked regression test) (087a8ea)

🐛 Bug fixes

• fix(verify-full): warm engine before scored checks (closes #352) (c595496)

📝 Documentation

• docs(tq3-mtp): add missing 04-gemma-vs-qwen.png chart (da9ef5e)
• docs: UPSTREAM Gemma4 TurboQuant row — exact config.py:101 mechanism + fix-PR set (fd8695f)
• docs+hygiene: track Gemma4 native-TurboQuant upstream blocker; gitignore new MoE cache dirs (f812715)
• docs(KERNEL_MATRIX): add Kernel Selection Philosophy section (287c766)

🧹 Other

• Merge PR #147: v0.8.0 — Universal pull (evaluate & serve any safetensors HF model) (#147 by @noonghunna)
• v0.8.0 [docs] PULL.md Quickstart (command-first, top-of-doc) + ARCHITECTURE one-liner: stage names are internal, users run one command (bef766d)
• v0.8.0 [review] pre-tag fixes: scrub internal-path leaks from shipped source + make .pull-captures-corpus tests CI-safe (skip-when-absent) (49d9bb4)
• v0.8.0 [docs] ARCHITECTURE.md: add the universal pull→gate→emit→loop pipeline to the mental model + scripts tree (current-state, was stale for v0.8.0) (1cdda19)
• v0.8.0 [UX] §7 two doc tracks: docs/PULL.md (user front-door) + docs/README.md (track spine) + README migration nudge (a0b3b5c)
• v0.8.0 [F] F8-fix: widen §6.1 Tier-1 OOM signature + pt3.actual regexes to real vLLM v0.21.0+ KV-cache-too-large phrasing — on-rig F8 caught classic-torch-only regexes miss the common KV-prediction failure (f92624d)
• v0.8.0 [F] F7: docs/LOOP.md contributor doc (Loop phase, grounded in shipped F1–F6) + CONTRACT-5(i) risk note (a8b30d6)
• v0.8.0 [F] F6: CONTRACT-5 mandatory content-hash kv_calc_version (G2) + G1 topo-verify + L2 fixture sync (1ac0481)
• v0.8.0 [F] F5: §6.3 canonical-tuple-hash dedup + bounded label scheme + collision-safe submit path (CONTRACT-4) (5de7224)
• v0.8.0 [F] F4: §6.2 inbound-trust pipeline raw→candidate→validated→Tier-1 + CONTRACT-3a derived-deferral (CONTRACT-3) (d758f08)
• v0.8.0 [F] F3: G6-A 3-part additive [E] touch (pt1.predicted_b_breakdown, pt3.failure_log_excerpt+actual, container-log capture) + §6.1 Tier-1 (CONTRACT-2) (b100979)
• v0.8.0 [F] F2: §6.1 Tier-2 semantic-fingerprint classifier + Appendix A seed DB (CONTRACT-2 Tier-2) (9f80d29)
• v0.8.0 [F] F1: FInput capture-bundle reader + schema-1 validation + key-normalization (CONTRACT-1) (1491cbc)
• v0.8.0 [E] E-outcome-fix: honest 3-state manifest outcome (partial-success != failed) — §6.2 partial is a capability-scoped success (71148d6)
• v0.8.0 [E] E3/E4-fix: boot lifecycle as context manager (server stays up for smoke+capture, teardown on ctx-exit) — on-rig E5 caught teardown-in-finally-before-smoke (f7c405a)
• v0.8.0 [E] E3-fix: smoke probes the real served-model-name (not literal "derived") + capture failure detail — on-rig E5 caught red-smoke-on-healthy-boot (16a1e4d)
• v0.8.0 [E] E2-fix-2: verify *.safetensors against HF API lfs.sha256 (not Xet-redirect-fragile HEAD x-linked-etag) — on-rig E5 caught false no-etag (3ae74bf)
• v0.8.0 [E] E2-fix: download via hf CLI subprocess (not huggingface_hub lib-import) — on-rig E5 caught ModuleNotFoundError (806a298)
• v0.8.0 [E] E5(docs): docs/PULL_EMIT_DERIVED.md (+ private ledger/recon-checklist updates) (d134d5a)
• v0.8.0 [E] E4: post-[C1] derived-[E] orchestration + trigger semantics + override force-capture (pt5) (2ed18aa)
• v0.8.0 [E] E3: derived boot (HF_HOME mount) + 4 §6 capture emitters + manifest + derived smoke floor (f327887)
• v0.8.0 [E] E2: HF download stage (download_set allowlist + x-linked-etag SHA, no-etag fail-closed, atomic staging) (7a2ec86)
• v0.8.0 [E] E1: generate_from_profile + derived-vllm template + EInput + CONTRACT-5 gate (411c84f)
• v0.8.0 Pull-Gate P5: docs/PULL_GATE.md (two-path model, 6-stratum taxonomy, §4.1 [C1], hardware-SM) (2582438)
• v0.8.0 Pull-Gate P4: stratum-5 + [C1] §4.1 total fn + stratum-6 [D] dry-run + pull orchestrator + exhaustive test-pull.sh (adf7a3b)
• v0.8.0 Pull-Gate P3: stratum-2 precondition + [C0] engine-support/runtime/hardware gate + [C2a] disk (4a1d385)
• v0.8.0 Pull-Gate P2: transformers deriver + ModelProfile/confidence + variant-scoped hf_repos schema (818b79c)
• v0.8.0 Pull-Gate P1: kv-calc generic-dense family + eligibility predicate + raw_verdict adapter (1bafcfe)
• v0.8.0: doc generated composes are not relocatable (run with --project-directory) (a2fc05e)
• v0.8.0 STEP 5: COMPOSE_GENERATOR.md + PATCH_POLICY.md (#141 contributor contract) (9546f99)
• v0.8.0 STEP 3+4: compose generator + 5-triple golden-parity test (#141) (6d7a043)
• v0.8.0 STEP 2: extract patch_attribution.py (sound body-only reaches(), test imports it) (60f3983)
• v0.8.0 Phase A-prime: enrich patch/profile data for #141 generator (compose_service_template, genesis_equipped, delivery metadata, drift_guards, drafter/model_slug/trc fold-ins) (9f23736)
• Add v0.8 Phase A patch attribution data (91a9622)

[Pin: git checkout v0.8.0] · Full diff