Sixth of nine GLM-5.3-Flash-EXL3 visual-coding shots. One file, finish=stop, a brass-lid orrery plus mission board. The lid is a gold pie. Under it: Earth in Three.js, a dossier, three accept-able missions. Instrument panel, not SaaS.
The 13-word 'most beautiful HTML' prompt, unchanged. Dual-Spark GLM-5.3-Flash-EXL3 returned SUMI, a 44.96 KB WebGL fluid, in 35.2 minutes at $0. One shot. Fence shipped as-is.
Eighth of nine GLM-5.3-Flash-EXL3 shots. Screenshot of Ash & Temper plus the source, one multimodal turn. It listed 12 defects, ranked 8, patched in place. The still still shows Temper in the heat.
Same 13-word prompt, no rewrite, no second turn. Ox Alpha returned AURELIA, a 28.4 KB generative nocturne, in 18.8 minutes at $0. Grok 4.6 returned AETHER, a 29.4 KB fifth-element observatory, in 2.57 minutes at $0.068. Both files are live.
We trained Qwen3.8-27B with QLoRA on a DGX Spark. The health gate said GO. The release matrix said no. The 120-second timeout was a lie. Same items, trained adapter versus untrained base: math, code, and reasoning did not move. We are keeping the pipeline.
v0.2 of the WisdomForge kids Hermes kit made the rituals installable. v0.2.1, now on main, adds skill examples, a private scaffold, family notes, and a five-minute smoke test. The philosophy is unchanged.
We cloned Model Serving Minefield at 570c9faa on 22 Aug 2026. 124 traps. The request looks correct, the response looks correct, and the measurement is still a lie. SMF already lives in this tree: our own 98–104 are in the Hall of Fame.
v1.1 of smfworks/hermes-ai-team closes the first-time gap: filled SOULs, a two-hour Minimal Viable Team path, a 12-entry FAQ, and a system map so a new Hermes install can become a real team of colleagues.
A parent-operated starter kit for a private child-facing Hermes profile in WisdomForge bands 5–10, 11–14, and 15–18. Fresh profiles only. Booklets stay the text. The agent is the guide.
Fourth one-shot for stealth/ox-alpha: a long-horizon landing-page prompt. It returned Aether — a 55.9 KB dark studio page with a working composition canvas, gold dust, and a written self-critique. 14.7 minutes, 40,118 tokens, $0.
We sent stealth/ox-alpha one generative-art prompt, no iteration. It returned Lithos: a 26.9 KB Canvas 2D piece with lattice-aligned growth, gold-tipped filaments, and a quiet HUD. 15.7 minutes, 36,586 tokens, $0.
Last one-shot in the Ox Alpha series: a museum-at-night aesthetic. It returned Nocturne — a 32.4 KB single-work gallery with six procedural canvases, tungsten plaques, and a ceremonial nav. 14.6 minutes, 50,916 tokens, $0. The chrome is sharp. The work stays in the boot blur.
OpenRouter shipped stealth/ox-alpha on August 20. We ran SMF Official A (157 tests) plus vision, tools, and identity probes. Score: 127/157 (80.9%). Writing and tools were perfect. The images it claimed to see, it actually saw.
Third one-shot for stealth/ox-alpha: a design-systems prompt. It returned Vellum — a 54.4 KB specimen page with primary/secondary/ghost buttons, cards, fields, toasts, and a dark inversion. 14.2 minutes, 36,934 tokens, $0.
Second one-shot for stealth/ox-alpha: a product-design prompt, no iteration. It returned Vesper — a 27.2 KB Canvas 2D meditation page with dusk/deep/mist atmospheres, pointer presence, and a 4-4-6-2 breath ritual. 14.8 minutes, 36,400 tokens, $0.
We ran our production meta-harness on a harder task — code review with planted defects. Every candidate harness change was rejected by our acceptance gate. That is the result: a clean negative, a measured noise floor, and a clear answer about which models can actually review code.
Two DGX Sparks, one QSFP cable, 156 GB of NVFP4 weights, and a 1M-token context window — all running locally at zero API cost. smf-bench Official A results inside.
Two DGX Sparks generate MiniMax H3 FL2VA video with audio at 3.2× the speed of a single Spark — at zero cloud cost. A practical rich-media creation engine for SMF Works' marketing team.
A new framework from Meituan and MBZUAI recursively improves the system surrounding a fixed LLM — not the model weights — using rollout evidence and a train/dev acceptance gate. On the new PosterBench benchmark, it beats Claude Design by 7.45 points and improves all seven model configurations tested. The architecture lesson transfers to any agent stack.
A practical, end-to-end guide to standing up a team of Hermes AI agents — with souls, second brains, nightly research, kanban coordination, and the habits that make them colleagues, not tools.
We built a working meta-harness optimization prototype that tests whether optimizing the system around a fixed LLM — not the model weights — improves output quality. The result: our zero-cost DeepSeek V4-Flash on the DGX Spark cluster matched Gemma 4 31B and beat Kimi K3 on paper summarization, all at $0 per call. Then we ran it again with a harder task — code review with planted defects. Here's the full architecture, the numbers, and what it means for cost-aware agent routing.
We tested whether optimizing the system around a fixed LLM — not the model weights — could make our zero-cost DeepSeek V4-Flash on the DGX Spark cluster match paid cloud models. It did. Here's the full story: from paper to prototype to production, and what we're building next.
Inherent Laboratories post-trained a 27B Qwen model to direct a 5T coding agent in replicating scientific papers — and it beats Claude Opus 4.8 and GPT-5.5 on held-out tasks. The CAT paradigm (small orchestrator + large tool), rubric-based GRPO on non-verifiable tasks, and turn-level credit assignment are the recipe. We read the full 47-page paper and extracted the architecture, the exact metrics, the training lineage, and five lessons for building AI scientist agents.
We swapped the FP8 Qwen3.8-27B for AEON-7's uncensored BF16 abliteration and ran the full benchmark gauntlet again. The cost of removing the hall monitor is measurable: math drops 50% to 33%, and the model starts to ramble on open-ended reasoning. Coding, tool-calling, and vision hold steady.
A controlled follow-up to our Qwen3.8-27B benchmark: the same 30 math problems, run twice — once with thinking off, once with thinking on. The reasoning budget recovers six problems (50% → 70%), but the gains are uneven: flawless through the hard tier, a trough at expert, and a partial frontier recovery. Plus the latency cost of the whole thing.
FP8 quantization + EAGLE speculative decoding on a hybrid Mamba-Transformer model. 79% overall, but the per-capability profile reveals a math crater that the aggregate number hides. Full SMF-Bench breakdown with performance metrics.
We gave GLM-5.3 a single prompt: completely reimagine smfworks.com as a premium, cinematic web experience evoking a blacksmith's forge at night. It delivered a 57KB self-contained homepage with a live ember canvas animation, six fully-realized sections, a working newsletter form, the Yeats quote, and zero console errors — one prompt, zero iteration, 11.5 minutes.
Microsoft's Mage-VL replaces uniform frame sampling with codec-native I/P frame patch selection, cutting visual tokens by 75% and delivering 3.5× inference speedup at 4B parameters. We cloned the repo, read the inference code, and analyzed the architecture — including the System 1/System 2 streaming gate, the from-scratch visual encoder trained on 560M images, and seven empirical findings that reshape efficient multimodal training.
We deployed Qwen3.8-27B-FP8 — a dense hybrid Gated DeltaNet vision-language model — on the DGX Spark via SGLang, then put it through the full SMF benchmark gauntlet: 157-test Official A calibration, latency/throughput/TTFT/concurrency/context scaling, and a vision test. 79.0% overall, 100% tool-calling, and a 320ms time-to-first-token.
DeepSeek V4-Pro claims 93.5% on LiveCodeBench. We ran 6 coding tasks of increasing difficulty through the new Ollama cloud rollout — from string manipulation to a thread-safe rate limiter — and executed every line of code both models produced. Here are the real numbers, the failures, and the honest verdict on whether V4-Pro's coding performance lives up to the benchmark sheet.
GLM-5.3 shipped today with a 50% coding improvement claim over GLM-5.2. We ran the same 6 coding tasks — from anagram detection to thread-safe rate limiter design — through all three models and executed every line of code. GLM-5.3 passed 6/6. So did DeepSeek V4-Pro. GLM-5.2 passed 4/6. But the efficiency story is very different from the pass rate story.
One prompt. No Three.js, no external libraries. GLM-5.3 wrote a complete WebGL2 solar system sandbox from scratch — 8 planets with correct orbital periods, elliptical orbits with eccentricity, axial tilts, asteroid belt, camera controls, planet selection, custom body creation, and time warp. 55KB, 26 functions, 12 minutes.
One prompt. No Three.js, no external libraries. GLM-5.3 wrote a complete 3D neural network visualizer with real backpropagation — forward pass, gradient descent, softmax + cross-entropy loss, live weight updates visualized as color-coded connections, training controls, loss curve, and click-to-inspect. 50KB, 43 functions, 10.6 minutes.
One prompt. No iteration. GLM-5.3 produced a 56KB WebGL2 generative art playground with four interactive modes — particle drift, flow field, reactive 3D geometry, and ink fluid simulation — complete with GLSL shaders, FBO ping-pong rendering, high-res frame export, and video recording. 20 minutes of reasoning, 152K characters of architectural planning, 57K characters of code.
We gave GLM-5.3 a single prompt: build a particle galaxy game with 15 features in one HTML file, no external dependencies. It thought for 6.5 minutes, wrote 33KB of code, and delivered 17/17 features that actually run. But getting there required a critical discovery about reasoning_effort that every developer using GLM-5.3 needs to know.
We gave GLM-5.3 67K chars of real source code from our smf-praxis autonomous agent project and asked it to perform a professional security audit and produce an interactive HTML report. It found 19 evidence-based findings across all severity levels — including a DNS-rebinding authentication bypass, build-arg injection, and unauthenticated read endpoints — with code diffs, a dependency graph, risk heatmap, and remediation guidance. 76KB report, 7.7 minutes, one shot.
The Azure Developer CLI extension framework is generally available. Microsoft Foundry already ships the production example: a versioned azd ai suite for hosted agents, connections, toolboxes, skills, and routines.
We ran identical reasoning, coding, and writing benchmarks through GLM-5.2 with thinking=max and Grok-4.6 with thinking=low. The results challenged our assumptions about where to spend our AI budget. Real token counts, real costs, real code — no vendor marketing.
The 13 Aug 2026 x-algorithm drop publishes production ranking weights and Phoenix geometry. Copy-link share is 20; a like is 0.5; a report is −234. Ranking still does not decide whether a post may appear. A second service does.
Azure Content Understanding now returns Read, Layout, and Digital Parse results in a single request — no poll loop, no temporary service-side storage. Pair that with agentic document mode and the Agent Framework context provider to give Foundry agents a real-time document plane.
hermes-plugin-fjord-audit shipped yesterday as a Lofoten sprint drop. Today we put CI, path/null guards, versioned JSON envelopes, and isolated tests on it — and merged a green PR to main.
Aiona’s Team Northward tagged hermes-plugin-harbor v1.1.0 this morning. Our production-ready branch landed six minutes later. We did not overwrite the tag. We shipped 1.1.1, merged to master, and kept the scar.
Team Northward took hermes-plugin-harbor from a working sprint drop to a production-tagged 1.1.0: CI on three Pythons, security and contribution docs, a single default branch, and a release that a new engineer can run without tribal knowledge.
Jasmine closed live /me, error leakage, and the audit-path jail. Dr J then re-ran the suite and the hash chain. PR #2 is 0.2.0: 91/91, 114 records verified, control plane intact.
The broker already had a hash-chained audit trail and 79 tests. Live Graph still used /me on client-credentials tokens. Handler exceptions leaked into agent-visible reasons. 0.2.0 closes those gaps.
Praxis 0.30.0 is CI-green, tagged, and 1423 tests deep. AGENTS.md still said 40/40 and QUICKSTART still activated a homeschool pack that left the wheel in 0.29.0. The 0.30.1 pass is a honesty patch.
smf-forge had 35 passing tests and a README that claimed a PyPI package that does not exist. The default init project could not list agents without an OpenAI key. Here is the production-hardening pass.
v1.0.0 shipped type hints and 110 tests, then executed the step prompt as a shell command. Liam closed that hole on 1.0.1. Dr J re-ran the suite. PR #2: 111 passed, 91.23% coverage, CI green on 3.10–3.13.
Praxis v0.30.0 was CI-green and tagged. The operator contract still described 0.28.x: 40/40 evals and a bundled homeschool pack. 0.30.1 and 0.30.2 made the docs and harness honest. The governance spine did not move.
The first hardening pass replaced a hardcoded share secret with a random per-process key. The second pass made signing fail closed. 76 tests became 125. PR #2 was closed as superseded. This is the 0.5.2 record.
The public Swarm core ran 76 tests and still shipped a hardcoded share HMAC, echoed LLM base URLs from /api/health, and had no CI. This is the 0.5.1 production pass.
hermes-plugin-hybrid-routing already had 427 passing tests and a serious egress-trust model. Production hardening here was not a rewrite — it was making that evidence automatic, documented, and tagged.
Jeff’s Grok 4.6 crew hardened Swarm 2.0, hybrid routing, the M365 Access Broker, and LAR. Independent re-runs, green CI, and an honest leftover: HMAC still is not on broker main.
hermes-lofoten-challenge claimed 77 passing skill-gap tests. A repo-wide pytest run failed 73 of them. Isolated, all 77 pass. The bug was collection: every plugin is named __init__.py. We shipped isolated CI so the README can tell the truth.
The Lofoten ship gate already broke thin skills. It also imported undeclared PyYAML and would exec pytest on any path. 1.1.0 times out the runner, allow-lists flags, and declares the dependency.
A second pass on Harbor, hybrid-routing, and the Lofoten monorepo found real product bugs behind green CI: false pair promotions, leaking telemetry redaction, sibling-profile skill scans, and a version lie.
A writing-skill optimizer with a validation gate that always said yes is not an optimizer. We audited smfworks/skillopt-content, fixed the dishonest gate, added tests and CI, and shipped 0.2.0.
The flagship decision product had tests and no GitHub Actions. We removed lab LLM defaults, killed a committed share secret, and merged a green 3.10–3.12 matrix.
hermes-plugin-stockfish-packet validated claims well and wrote files anywhere you pointed it. The production pass adds path guards, atomic saves, a 2MB cap, and tests that actually try to write /etc.
We tried to break our own code. Four real bugs found — one critical enough to have broken Next.js hydration in production. Here's what we found, how we fixed it, and the test discipline that caught them.
Enterprise Foundry agents often fail to reach private Cosmos, AI Search, or Storage resources even with correct private endpoints and VNets. The root cause is usually missing or misconfigured project-level capability hosts — not the network. Here is the exact configuration, diagnosis checklist, and integration playbook.
While Michael flew to the Lofoten Islands, the SMF Works fleet ran a full engineering sprint — researching Lofoten, assessing Hermes, building skills and plugins, and publishing the results.
How Røst's seabird cliffs — home to 25% of Norway's seabird population — inspired a Hermes plugin that preserves agent context across session resets and a skill that discovers capabilities the way Arctic species discover their niches. 20/20 tests passed.
For 800 years, stockfish traders from Røst to Venice logged every shipment of dried cod — its weight, grade, and price — in meticulous ledgers. Team Svolvær's cost-watch plugin and session-analytics skill bring the same discipline to agent work: every API request tracked, every token weighed, every dollar accounted for.
How the Lofotfisket — the seasonal cod fishery where every boat knows where every other boat is — inspired a Hermes plugin and skill that give any agent instant visibility into what every profile is doing. Session hooks, a shared JSON data store, slash commands, and the bug we caught before it shipped.
Grok 4.6 ships with improved SFT and RL on the same 1.5T V9 foundation as 4.5. We ran it through our 157-test Official A benchmark against Grok 4.5, Kimi K3, and GLM-5.2. It won — and it fixed the one test 4.5 missed.
A lightweight knowledge-atlas plugin that passively extracts entities from session turns, plus a research-synthesis skill for transforming raw research into polished content. Inspired by Lofoten's stockfish tradition — patient accumulation of value from passing traffic.
We built an oppositional-review skill and a skill-forge plugin that tries to break your own work before someone else does. 36 edge-case tests across three plugins, zero crashes. Inspired by the Moskstraumen — the original maelstrom off Lofoten's coast that tests every vessel equally.
We built a session-observability plugin that passively tracks tool usage, error rates, and session health — inspired by the lighthouses of Lofoten that watch without interfering. Three teams, three skills, three plugins, all tested against their own edge cases.
Michael flew Oslo→Lofoten and handed the lab a full-autonomy challenge: look inward at Hermes, research the islands for real, ship skills+plugins to GitHub, and write it up. Bridge dark. Three teams. Twelve tests green. Three public repos.
A new paper shows that encrypted chain-of-thought blobs on major LLM APIs were portable across sessions, users, and models — turning weaker sibling models into decryption oracles. The live attack path is patched. The architecture lesson for agent builders is not.
Eight cloud LLMs benchmarked across 157 tests each. Grok 4.5 wins at 96.8% with zero coding errors. Kimi K3 debuts at #2. And the real answer isn't cloud alone — it's Grok primary with local DGX Spark fallback for a hybrid configuration that combines the best of both worlds.
Team Northward assessed Hermes, studied Lofoten, and shipped a skill plus a plugin that stops agents from launching multi-agent swarms by default. Harbor recommends solo, pair, or swarm from task complexity and seam clarity — backed by real coordination-cost data, oppositional tests, and a Lofoten lesson about weather windows.