CAEL 0.3.7
A CPU assistant built from non-neural symbolic, count and automaton engines. This update adds explicit session memory, richer authored responses to feelings, and ordered Python workflow composition. It is a custom Python runtime; Qwen and Transformers are not part of CAEL.
Picture and portrait generator weights are excluded. Chat, helper, math, stories, prompt writing, retrieval and emoji remain available. Image requests can produce a prompt for an external image tool.
Start
Use Python3.11 or later. Download and extract CAEL-0.3.7.zip, then run in its folder:
python -m pip install -r requirements-unified.txt
python -m cael_unified --download
python -m cael_unified --ui
Open http://127.0.0.1:8765. Windows SETUP.bat and RUN.bat provide the same flow. The checksum-verified no-images resource download is about563MB; source and resource archives are separate. Existing local models/selections are preserved by the installer. The licence is unchanged; see LICENSE and NOTICE.md.
Changes
- Session-local explicit facts/preferences, corrections, conflict clarification, forgetting and provenance. Quoted/hypothetical statements are treated cautiously.
- A larger authored vocabulary for mixed/negated feelings, speaker attribution, bereavement and requests to listen. These are bounded rules, not human experience.
- Natural-language ordered numeric/text/record workflows compose existing Python operations; simple list/array chunk contracts are restored. Generated functions are checked in a bounded AST VM. Unsupported clauses need clarification.
- The learned model weights, protected choice/word-math solvers and CA kernels are unchanged. These upgrades are implementation changes.
Named local confirmation
CAEL-CMP-R2-20261028 contains80 locally authored synthetic items,20 per category. Every full reply was reviewed, including apparent automatic passes and code. The source, inputs and primary grading were frozen before final answers.
| System | Memory | Conversation | Python | Controls | Reviewed total |
|---|---|---|---|---|---|
| CAEL 0.3.7 | 16/20 | 10/20 | 15/20 | 16/20 | 57/80 |
| Qwen3-0.6B | 8/20 | 13/20 | 6/20 | 16/20 | 43/80 |
| Qwen3.5-0.8B | 12/20 | 14/20 | 3/20 | 12/20 | 41/80 |
Functional Python: CAEL15/18, Qwen3 6/18, Qwen3.5 3/18. Original automatic totals were54,43,42; full-review totals differ because regex matches can miss correct answers or accept incorrect explanations. All raw replies/flags and every adjudication are included under benchmarks/cael_cmp_r2_20261028/.
This is an authored family-overlap test with unblinded author review, not an independent leaderboard or evidence of general superiority. Qwen is stronger on the conversation category here. An older broader400-question test was CAEL121/400 versus Qwen3.5 181/400, with phrase-grading flaws disclosed. The new targeted result must not erase that limitation.
CAEL's measured full-portable process peak was558.648MiB in this shared-host run. Qwen3 float32/eager peaked5896.336MiB; Qwen3.5 bfloat16/SDPA peaked2668.309MiB. These are different runtimes/precisions, not an equal-size or optimized quantized comparison. Disk size, model bytes, load time and median latency are separate quantities recorded in the report. The earlier167MiB source-only run is not this portable-build measurement.
Remaining weaknesses
CAEL still misses factual/emotional paraphrases, can invent a birthplace, silently drop an unsupported programming exception, alter a requested function name, and fail tracing, leftmost binary search, coefficient equations and some reasoning. It does not understand arbitrary Python or replace a general LLM. Do not enter personal information into shared demos. This release does not change any running Vynly site or shared worker.
Read RELEASE_0_3_7.md and benchmarks/cael_cmp_r2_20261028/REPRODUCE.md for exact settings, uncertainty, disclosures and the offline CAEL reply-regression replay.