APEX-0
Trained, checkpointed, resumed, reloaded and used for local text generation. Lifecycle tests pass.
Capability qualification: NOT ESTABLISHED
Evidence ceiling E3Proof boundary
APEX Model & Research Status
A model exists. Its capability still has to be established.
APEX-0 is an executed local research lifecycle model. No general, industry, solution-building or frontier intelligence qualification is claimed.
APEX-0Research lifecycle tested
Qualified intelligence models0
APEX production admissionHOLD
Customer-liveNot established
Model lineage
Generations describe the program, not achieved intelligence.
APEX-1
No trained and qualified artifact is established for this generation.
Capability qualification: NOT ESTABLISHED
APEX-2
No trained and qualified artifact is established for this generation.
Capability qualification: NOT ESTABLISHED
APEX-3
No trained and qualified artifact is established for this generation.
Capability qualification: NOT ESTABLISHED
APEX-4
No trained and qualified artifact is established for this generation.
Capability qualification: NOT ESTABLISHED
APEX Frontier
No trained and qualified artifact is established for this generation.
Capability qualification: NOT ESTABLISHED
Measured result
Next-token loss is a lifecycle measurement.
- Measure
- NEXT_TOKEN_LOSS_ONLY
- Cross entropy
- 1.916830
- Perplexity
- 6.799369
- Capability established
- No
Seller-authored lifecycle corpus; next-token loss is not task capability or frontier qualification.
- 01Model exists
- 02Capability measured
- 03Baseline outperformed
- 04Result reproduced
- 05Research goal achieved
Only the research lifecycle and next-token measurement are established here. Baseline superiority, independent reproduction and frontier goals remain unestablished.
Implemented engineering
A controlled foundation for research.
Model Foundry
APEX-0 training, signed checkpoints, local reload and inference execute.
This is lifecycle validation, not capable general or frontier intelligence.
Inspect this layerImprovement Governor
Rights, immutable plans, checkpoint and final-evaluation locks constrain local improvement.
Successful self-improvement and qualified release promotion are not established.
Inspect this layerVersioned Intelligence Registry
Artifact admission, byte integrity, rights and revocation checks execute.
A registered artifact is not automatically capable or production-qualified.
Inspect this layerRollback Engine
Local checkpoint resume and committed-state restart/restore are exercised.
Protected global trust restore, distributed recovery and second-host portability remain open.
Inspect this layerResearch Preregistration
Immutable plans, negative results and distinct replication references persist.
A replication reference does not self-award independent reproduction or a causal result.
Inspect this layerActive research
Research questions remain open.
Contract Challenge
Bounded declared-predicate omission and challenge checks.
Active Evidence Acquisition
Rights-bound measurement requests; authoritative evidence remains external.
Proof-Bound Permissions
Host-enforced authority, trust, scope and resource boundaries.
Causal Outcome Modeling
A program objective with no qualified learned result or independent reproduction established.
Verified Invention
A program objective with no qualified learned result or independent reproduction established.
Federated Intelligence
A program objective with no qualified learned result or independent reproduction established.
Cross-Domain Transfer
A program objective with no qualified learned result or independent reproduction established.
Corrected-Outcome Learning
A program objective with no qualified learned result or independent reproduction established.
Adaptive Research Allocation
A program objective with no qualified learned result or independent reproduction established.
Capability Creation
A program objective with no qualified learned result or independent reproduction established.
Governed Self-Improvement
A program objective with no qualified learned result or independent reproduction established.
External validation
Outside evidence must come from outside.
No qualifying buyer-authored result, independent implementation, independent evaluation or external witness was obtained. Seller evidence remains within E1–E3.
Inspect existing verification researchLong-horizon objectives
General, domain and solution intelligence are research goals.
Capable first-party models, retained learning, useful invention and cross-domain transfer require substantive data, predeclared evaluations and authorized review. No superintelligence or universal industry mastery is claimed.