Registry · Models

Release records and source-attributed upstream snapshots.

LeanLogix owns local release evidence. External observations are legacy, dated snapshots whose upstream facts remain canonical in MOEModels; their signatures prove receipt integrity, not release authorization. A row or channel label does not prove hosting, endpoint reachability, loaded weights, or inference.

Open the studioVerify any model

13

Live registry evidence rows

1

Rows labeled production · not serving proof

10

Externally observed records

2

Evidence semantics · release vs observation

The registry

Local release records, separated from upstream snapshots

This is not a universal model registry or hosting inventory. MOEModels owns upstream discovery and public facts. LeanLogix retains source-attributed snapshots only for compatibility and local audit; serving still requires a separate authorized release, configured endpoint, and request-time gate.

Ecosystem models · 8

LockedIn Labs ecosystem models

The models built by the LockedIn Labs ecosystem and LeanLogix’s own platform — SprintLoop, ControlFrame, HealthNext, and the routing/infra layer. These are the builds the platform was first proven on.

ModelBaseParams · SizeVersionChannelEvalCompliancePassport
Portfolio Router
MLX LoRA · rank 16 · 16 layers · 400 iters cosine · promoted val-trough ckpt 200 (val 0.018)
portfolio-router-gold v1 (525 ex · 473 train / 52 valid · synthetic+public · no customer data) · Tiny fast task router: classifies each request to the right portfolio model + reason, in strict JSON. Base Qwen2.5-1.5B-Instruct (Apache-2.0). Held-out router eval 96/100 (base 76); refusal recall 5/5 on adversarial probes. Candidate — NOT certified/production.
Qwen2.5-1.5B-Instruct1.5B · 0.0422 GBv1Candidate96/100In review
HealthNext-Care-7B
MLX LoRA · rank 16 · 400 iters · cosine
HealthNext-Care corpus: 621 ex (589 train / 32 valid), 6 areas, 0 malformed/PHI/dupes · In-boundary healthcare ops reasoning. Base Qwen2.5-7B-Instruct (Apache-2.0). Val loss 3.671 to 1.108 (best). Serving base for per-customer forks. Domain hard-suite eval pending.
Qwen2.5-7B-Instruct7Bv126/100In review
HealthNext-Care-32B
MLX LoRA (QLoRA) · rank 16 · promoted val-trough ckpt 125 (val 0.920)
HealthNext-Care corpus: 621 ex (589 train / 32 valid), 6 areas, 0 malformed/PHI/dupes · Quality-tier candidate — wins the hard suite 34 vs 26 over Care-7B, decisive on governance (71.4% vs 39.3%). Promoted from val-loss trough, not last iter. Care-7B remains served default.
Qwen2.5-32B-Instruct32B · 32 GBv1-candidateCandidate34/100In review
SprintLoop-7B
LoRA · MLX (Apple Silicon)
sprintloop-corpus · 2,800 · no customer data · Governed coding + plan/verify; refuses unsafe ops. Served in production.
Qwen2.5-Coder-7B-Instruct7B · 1.4 GBv6Production89/100Artifact sealed · passport pending
HealthNext-7B
QLoRA · planned
public CMS corpora + synthetic (RAG; no PHI in weights) · Compliant healthcare ops — tuned on PUBLIC CMS rules; PHI stays in RAG/runtime, never in weights.
Qwen2.5-7B-Instruct7BplannedDevPending
SprintLoop-7B · Edge
GGUF Q4_K_M · quality retention 97%
derived from sprintloop-7b v3 · Lean build for on-device / CPU. Quality-retention report attached.
SprintLoop-7B v37B · 4.1 GBq4-v1Candidate80/100In review
ControlFrame-Compliance-7B
LoRA · planned
controlframe-gold (in prep) · Compliance-framework interpretation + evidence planning. First governed build.
Qwen2.5-7B-Instruct7BdraftDevPending
SprintLoop-32B
QLoRA · MLX (Apple Silicon)
sprintloop-corpus · 2,800 · no customer data · Scale tier — wins the hard benchmark 90 vs 83. Awaiting promotion.
Qwen2.5-Coder-32B-Instruct (8-bit)32B · 5 GBv1Candidate90/100In review

External records · 5

External and partner evidence

External, partner, and customer model records that LeanLogix does not own. Inclusion preserves evidence and review state; it does not claim the platform hosts or serves the model.

ModelBaseParams · SizeVersionChannelEvalCompliancePassport
healthnext-care-prod-demo
lora-fork
tenant:prod-demo corpus:ab12cd34 · Per-customer in-boundary fork for tenant "prod-demo" — trained on NON-sensitive environment only; PHI/secrets held out to runtime RAG.
healthnext-care-7blorav1In review
healthnext-care-acme-demo
lora-fork
tenant:acme-demo corpus:ab12cd34 · Per-customer in-boundary fork for tenant "acme-demo" — trained on NON-sensitive environment only; PHI/secrets held out to runtime RAG.
healthnext-care-7blorav1In review
healthnext-care-meridian-health
lora-fork
tenant:meridian-health corpus:1d8c8a28 · Per-customer in-boundary fork for tenant "meridian-health" — trained on NON-sensitive environment only; PHI/secrets held out to runtime RAG.
healthnext-care-7blorav1In review
ThirdCandle-Advisor-7B
QLoRA · planned
trade-discipline-narratives (in prep) · Evidence-only trading advisor narratives — no buy/sell directives or predictions; discipline + risk framing.
Qwen2.5-7B-Instruct7BplannedDevPending
Itkalam-7B
QLoRA · planned
itkalam-arabic-corpus (in prep) · Conversational Arabic tutor — MSA + Egyptian/Levantine/Gulf/Maghrebi dialects. First Itkalam build.
Qwen2.5-7B-Instruct7BplannedDevPending

A production channel is a registry fact, not proof that an endpoint is configured, reachable, or serving the recorded weights. Ecosystem and external records use the same evidence contract. The passport column reads passport-ready until explicit signed-passport evidence is present. A stored artifact and hash read artifact sealed — useful integrity evidence, but not a signed passport, certification, hosting claim, or Verify action. Early builds with no eval or evidence are shown exactly as they stand; the registry doesn't round drafts up to releases.

Observed · public-fact snapshots we do not host

A tamper-evident provenance index for widely used models

MOEModels owns upstream discovery and canonical public model facts. These legacy observation receipts are tamper-evident, dated snapshots of the source LeanLogix read for compatibility and local audit. Signature validity proves snapshot integrity under its signer; it does not authorize a release, certify the source, or create a benchmark or quality claim.

ModelParamsBase · contextLicenseObservedPassport
Llama 3.1 70B Instruct
Meta
decoder-only transformer (Grouped-Query Attention; not MoE)
70B128K tokensLlama 3.1 Community License Agreement2026-06-25integrity only
Qwen2.5-72B-Instruct
Alibaba (Qwen Team)
dense decoder-only transformer (RoPE, SwiGLU, RMSNorm, GQA, 80 layers)
72.7B total · 70.0B non-embedding32,768 native · 131,072 with YaRNQwen License (license_name: qwen)2026-06-25integrity only
Mistral Small 3 (Mistral-Small-24B-Instruct-2501)
Mistral AI
decoder-only transformer (dense — community-attested, not stated verbatim on the official card)
24B32K tokensApache-2.02026-06-25integrity only
DeepSeek-V3
DeepSeek-AI
Mixture-of-Experts (Multi-head Latent Attention, DeepSeekMoE, Multi-Token Prediction)
671B total · 37B activated per token128K tokensDeepSeek License Agreement v1.0 (weights) · MIT (code)2026-06-25integrity only
GPT-4o
OpenAI
single end-to-end multimodal network (internal architecture not publicly disclosed)
not publicly disclosed128K tokens (16,384 max output)proprietary (OpenAI API / commercial terms)2026-06-25integrity only
Qwen2.5-Coder-7B-Instruct
Alibaba (Qwen Team)
dense decoder-only transformer (RoPE, SwiGLU, RMSNorm, GQA, 28 layers)
7.61B total · 6.53B non-embedding32,768 native · 131,072 with YaRNApache-2.02026-06-25integrity only
Llama 3.3 70B Instruct
Meta
decoder-only transformer (Grouped-Query Attention; not MoE)
70B128K tokensLlama 3.3 Community License Agreement2026-06-25integrity only
Gemma 2 27B Instruction-Tuned
Google DeepMind
decoder-only transformer (interleaved local/global attention, logit soft-capping, GQA)
27.2B8,192 tokensGemma Terms of Use2026-06-25integrity only
Phi-4
Microsoft Research
dense decoder-only transformer (trained heavily on synthetic + filtered data)
14B16,384 tokensMIT2026-06-25integrity only
Mixtral 8x22B Instruct v0.1
Mistral AI
Sparse Mixture-of-Experts (8 experts, top-2 routing)
141B total · 39B active per token64K tokensApache-2.02026-06-25integrity only

All 10 observation receipts had their embedded-key signatures and observation schemas re-checked server-side as you loaded this page. These are observed, not hosted, evaluated, released, or served. A valid embedded-key signature proves snapshot integrity; LeanLogix authority is trusted only when the signer matches an independently configured pin. Open any row to inspect both results, or paste the receipt into the public verifier at lockedinlabs.ai/verify. Embedded key a97122c37f…97fac1e1.

Repository sandbox

Illustrative dataset records

These static seed rows demonstrate the intended dataset contract. They are not a live registry read, do not establish that a scan ran in production, and are kept separate from release authority.

DatasetRowsStatusPII scanCustomer data
sprintloop-corpus2,800 verifiedapprovedclean · 0 findingsNo
controlframe-goldin prepregisteredpendingNo

The governed library

This catalog is the public read of the registry

The operator workspace exposes sandbox examples separately from live version history, evaluation records, release state, and authenticated passport evidence. A seed or channel label never establishes serving eligibility.

Inspect the registry behind these evidence records

13 model records on the public catalog today — 8 ecosystem records and 5 external records. Step into the Studio to inspect evidence and release state; use the governed gateway to determine actual serving eligibility.

Open the studioHow the registry works