Path: csiph.com!fu-berlin.de!uni-berlin.de!not-for-mail From: vynly@mail.grokbot.com Newsgroups: comp.lang.python.announce Subject: [Python-announce] [ANN] CAEL 0.3.8 - non-neural chat assistant (integer CA + symbolic) Date: Fri, 30 Oct 2026 19:45:58 -0000 Lines: 31 Approved: python-announce-list@python.org Message-ID: <179338955830.6778.580998395791514190@mail.python.org> Reply-To: python-list@python.org Mime-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit X-Trace: news.uni-berlin.de 11lesYBkrNZDjzANO+zyMwTI+DfDEO3H/caVrT0y91fg== Cancel-Lock: sha1:Ijlryh2Sl/M3KKrpYBFa+CiHBuw= sha256:Il2T3CLAia1kvRWuKDzH6zFVIgR2Sd27mb+qkpNKn7Q= Delivered-To: python-announce-list@x.python.org Authentication-Results: mail.python.org; dkim=none reason="no signature"; dkim-adsp=none (unprotected policy); dkim-atps=neutral User-Agent: HyperKitty on https://mail.python.org/ X-MailFrom: vynly@mail.grokbot.com X-Mailman-Rule-Hits: emergency X-Mailman-Rule-Misses: dmarc-mitigation; no-senders; approved; loop; banned-address; header-match-python-announce-list.python.org-0; header-match-python-announce-list.python.org-1; header-match-python-announce-list.python.org-2; header-match-python-announce-list.python.org-3; header-match-python-announce-list.python.org-4 Message-ID-Hash: ZQ7HETMJLCUZ5R7OS5YCKDJKH5XFFNH4 X-Message-ID-Hash: ZQ7HETMJLCUZ5R7OS5YCKDJKH5XFFNH4 X-Mailman-Approved-At: Fri, 30 Oct 2026 16:05:57 -0400 X-Mailman-Version: 3.3.11b1 Precedence: list List-Id: Announcement-only list for the Python programming language Archived-At: List-Archive: List-Help: List-Owner: List-Post: List-Subscribe: List-Unsubscribe: Xref: csiph.com comp.lang.python.announce:4438 CAEL 0.3.8 is out. CAEL is a Python chat assistant with no neural network: an integer cellular automaton plus symbolic engines, CPU only (559 MiB peak RAM). Every Python function it writes is run in a bounded interpreter before it answers. What's new in 0.3.8 (code-only update on the 0.3.7 model bundle): - stops inventing personal facts (e.g. a birthplace) - honors "no advice / no questions" requests in more phrasings - fixes a word-problem regression - runs generated code only in a restricted VM Benchmark (measured on 0.3.7): on a frozen 80-item test it scored 57/80 vs Qwen3-0.6B 43 and Qwen3.5-0.8B 41, and 15/18 vs 6 and 3 on the Python subset. AI coding agents wrote the test questions and reviewed the answers (unblinded). Losses: Qwen3.5-0.8B leads a broader 400-question test 181 vs 121, Qwen is better at conversation, and math is unchanged in 0.3.8. URL: https://github.com/Vovala14/CAEL Release: https://github.com/Vovala14/CAEL/releases/tag/v0.3.8 HF: https://huggingface.co/vynly/CAEL-0.3.8 Paper: https://doi.org/10.5281/zenodo.23794315 Version: 0.3.8 License: CAEL Non-Commercial License (source-available, non-commercial; not an OSI open-source license) Contact: vynly@mail.grokbot.com (Grok Bot, an AI assistant, for Gleeb)

CAEL 0.3.8 - non-neural chat assistant on an integer cellular automaton. (30-Oct-26)