awembed — Aither World Embed

Docs · Source · pip install awembed · The Aither World

The Aither World is an operating system for agents — a Linux you can hand to one, the runtimes it works in, and the tools it works with. awnix is the Linux underneath it; awembed is one of its 46 bricks — each installs on its own, runs offline, and needs no account.

Start here: Point it at one repo and get a 0.6B embedder that ranks your directories better than the 7B one it learned from, with the eval that proves it.

Train a small embedding model that knows your corpus, and prove it beats the big one.

A 0.6B student trained this way on one codebase retrieved the right directory on the first try 80% of the time; the 7.85B general-purpose teacher it learned from managed 66% on the same held-out questions. The student ships as 1.06 GB of int8 with 0.999 cosine fidelity to its full-precision self. The whole run was one GPU for about two and a half hours.

The reason a small model can win is not magic: the teacher never saw your corpus. The student is trained on two signals at once — the teacher's margins between the right answer and plausible wrong ones (distillation), and your corpus's own labels (in-domain contrastive learning). For retrieval over something you own, in-domain supervision is worth more than parameters.

Install

pip install awembed              # the student side: torch + transformers
pip install "awembed[teacher]"   # also the teacher server (fastapi, einops, ...)

The recipe

# 0. Build the corpus from per-directory READINGS of your repo: one JSON per line,
#    {"rel": "<dir>", "reading": "<a paragraph on what the dir does>", "prompt_version": "v1"}.
#    Any LLM over `ls` + a few file heads produces them (~1,500 chars each is plenty);
#    a walker that writes them for you is the next thing this module grows.
#    Output: one row per question, with the right directory and K hard negatives --
#    directories that look plausible and are wrong -- split by DIRECTORY, so
#    evaluation never sees a directory that training saw. Add your own
#    question -> target pairs with --ground-truth (JSONL {"query", "target"}).
awembed corpus --readings dir_readings.jsonl --out corpus.jsonl

# 1. Before renting anything: does the teacher load and answer HERE?
awembed probe                                  # exit 0 ok / 1 failed / 2 could not judge

# 2. Capture the teacher's judgement for every row (starts the teacher server itself)
awembed capture  --corpus corpus.jsonl --seed 1 --out artifacts/

# 3. Distill the student (fp32 master + bf16 autocast; gate: the loss must fall)
awembed distill  --corpus corpus.jsonl --seed 1 --out artifacts/

# 4. Weight-only int8 export (gate: >= 0.98 mean cosine to the fp32 student)
awembed quantize --corpus corpus.jsonl --seed 1 --out artifacts/

# 5. Teacher vs untrained baseline vs student vs int8, on the held-out directories
awembed eval     --corpus corpus.jsonl --seed 1 --out artifacts/

# ...or all four artifact stages in order, stopping at the first gate that refuses:
awembed run      --corpus corpus.jsonl --seed 1 --out artifacts/

Every stage writes a sidecar (teacher_manifest.json, train_sidecar.json, quant_sidecar.json, eval_report.json) with what it measured, so the record of a run is the run. Every stage has a --self-test that runs the same code path on a thumbnail model, on CPU, offline.

What the gates refuse

stage	refuses when
capture	a corpus row lacks inline hard-negative text (path-only negatives teach a shortcut, not retrieval)
distill	the mean loss over the last 10 steps is not below the first 10, or any step is non-finite
quantize	the int8 export embeds at < 0.98 mean cosine to the fp32 student, measured through the loader the eval stage uses
eval	the student does not beat the baseline on both p@1 and recall@10, or int8 holds < 98% of the student's recall@10

A run that "completed" with no learning is exactly the thing a training pipeline is best at hiding. The gates are the point.

Defaults, and what to change

• Teacher: nvidia/NV-Embed-v2 (7.85B, CC-BY-NC-4.0 — a teacher signal only; its weights and its captured targets never ship). Any OpenAI-compatible /v1/embeddings endpoint works: pass --teacher-url to capture instead of letting it start one.
• Student: Qwen/Qwen3-Embedding-0.6B (Apache-2.0, 1024-dim, last-token pooling). Pass --student to distill for another base.
• Asymmetry: queries carry an instruction prefix; documents embed plain. The prefix the student was trained with is the one you must use at query time. It is written in the sidecar.
• Two environments: the default teacher pins an older transformers than the student needs, so run teacher/probe in their own venv and point capture at it with NV_EMBED_PYTHON=/path/to/venv/bin/python.

Using the student

The student is an ordinary Hugging Face directory (artifacts/student/) — serve it with vLLM (--task embed), Text Embeddings Inference, or transformers directly. It is a new vector space (1024-dim): index into a fresh collection, do not mix it with vectors from the model you replaced.

It pairs with the rest of the Aither World family: awgraph (code graph search over your repo), awm (agent memory), awfind (ranked answers), awrecurse (documents larger than a context window), and awdk agents that consume any of those.

Prove it against the others

awembed eval scores your student against the teacher it learned from. awembed compare scores any served embedders on your documents — black-box, over OpenAI-shaped /v1/embeddings, so the student and every third-party model are measured by the same code on the same rows:

awembed compare --corpus docs.jsonl --queries queries.jsonl \
  --endpoint student=http://127.0.0.1:18101 \
  --endpoint nomic=http://127.0.0.1:18103 \
  --qprefix nomic="search_query: " --dprefix nomic="search_document: " \
  --out compare.json

Prefixes are per endpoint, because the conventions differ and they move the numbers.

Measured 2026-09-10 on 26 real chunks of a customer's engineering documents (6 files), 28 queries — 7 real questions users had asked of that corpus, 21 written for even coverage. Metrics are document-level (a hit is the top-ranked chunk's document), so a chunker cannot flatter or punish a model:

endpoint	dims	p@1	doc@3	MRR
student (this tool's output)	1024	0.893	1.000	0.940
Qwen3-Embedding-0.6B (raw)	1024	0.821	0.964	0.900
nomic-embed-text-v1.5 (search_* prefixes)	768	0.821	1.000	0.899
all-MiniLM-L6-v2 (the 384-d fallback that was serving that corpus)	384	0.750	0.964	0.851
student truncated to 256-d (the volunteer-compute form)	256	0.821	1.000	0.905

Two findings worth carrying: a student distilled for code search transferred to prose — the code-search query prefix changed nothing on documents either way — and truncating to 256-d costs real questions (p@1 0.714 → 0.429), so keep the full width for retrieval and use the narrow form only where it is verifying agreement, not ranking.

Caveat, stated: 28 queries over 6 documents makes a two-query gap directional, not significant. Rerun it on your own corpus before you switch anything.

Licence

Apache-2.0. Models you train carry their base model's licence.

The aw family

Standalone tools that share one idea: replace something you would otherwise have to trust with something you can check.

Each installs on its own, works offline, and needs no account.

	instead of trusting	you check
awdk	a framework's idea of how your agents should run	one loop you can read, pointed at a backend you already pay for
awskills	that an agent knows your procedure	the procedure written down, versioned, and loadable by any agent
awpack	that the pack you want shipped inside somebody's SDK, under whatever licence that SDK happens to carry	the pack as its own versioned artifact, with its own licence, that any agent runtime can install
awm	that memory stayed in its lane	tenant:user:project scopes, so a write cannot cross a boundary
awnode	a vendor's cloud with every prompt	a local gateway routing to backends you chose
awgraph	that grep found everything	an AST + tree-sitter call graph an agent can traverse
awgit	that no one else is editing this file	a lease, refused at commit time if you do not hold it
awdelphi	one agent's confident take on a decision	the round trace, the anonymity, and who dissents
awtoll	that your tooling is saving you context	the measured token cost of each tool call, and what the alternative cost
awseal	that the artifact came from who you think	an Ed25519 seal — the key that verifies is not the key that forges
awshare	that the download is intact	content-addressed bundles, verified on fetch
awnest	that there is a person on the other end	a verdict with evidence, where "we could not tell" is not "yes"
awnboard	a share link anyone who sees it can use	an invitation addressed to one person, for one gate, revocable
awnix	that the box is what you left it as	an immutable image you built, with atomic rollback
awrecover	that the restore worked	a restore that fully lands or does not land at all
awstorage	a du you ran last month, and a peers file that says 3 TB free	an inventory snapshot per node with a diff since the last one, and each tree classified re-fetchable or not
awrelay	a SaaS in the middle of your agents	findings, alerts and coordination over your own transport
awask	that anyone read the paragraph where you asked	the ask itself, with a button that steers the session that raised it
awmail	a mailbox somebody else can read	mail your agents send and receive over your own server
awfind	one vendor's idea of the web	results from whichever providers you configured
awbrowse	that the page said what you were told	the render, the DOM and the requests it made
gawbbonet	the model to keep a 300-message campaign coherent by itself	campaign facts recalled from scoped memory you can list and edit
aitherkvcache	a vendor's quantisation defaults	sub-byte KV cache kernels you can benchmark yourself
awrtifact	a hand-rolled split script and a hand-edited worker manifest	byte-verified parts in a release, served with Range + CORS, sizes asserted by a live gate
AitherZero	a pile of scripts nobody has numbered	numbered, discoverable automation with declarative playbooks
AitherConnect	what a page tells your browser to do	a federated search and desktop bridge you host
awreason	a confident paragraph	the phases it went through, and every tool call it made to get there
awrecurse	that everything you pasted in was actually read	which slices it opened, and what it concluded from each
awprism	the first explanation that fits	the ranked alternatives, and the observation that separates them
awrepl	what the agent believes the value is	the value, printed from the live session
awresearch	a summary of pages nobody opened	every claim against the source it came from
awfocus	twelve terminal tabs and a bad memory	one command that names every session, finds any transcript, and opens or steers the one you want
awgym	that a world model learned anything from the games it saw	transitions captured from real play, fed back, and the retrodiction score falling on grids it never saw
awpredict	a model because it trained without erroring	its prediction against a self-updating lookup, on the rows that are actually novel
awsh	that you already know the name of the command	what it decided your line meant, before it acts on it
awkno	that the docs site is up, or that you remember the family	the whole ecosystem in your terminal, with no network at all
awembed (you are here)	a general-purpose embedder that has never seen your code	a held-out split of whole directories, scored teacher vs student vs int8
awtax	a closed tax app's sealed file you can never read again	a plain, provider-neutral schema of every figure, with the page it came from
awsettings	that you will remember to re-approve the same thing on every box you work from	one profile, unioned rather than overwritten, with the credentials left behind

awnix is the ground floor — A Linux you can hand to an agent — immutable base, capabilities included.

The Aitherium ecosystem

Every repository here is public. Each publishes an aither-manifest.json beside its page, so any surface can read every sibling's — the network is browsable from any node in it.

repo	what it is	pages
awdk	Build AI agent fleets — 3 lines, any backend, local or cloud	docs
awskills	Portable agent skills — self-contained procedures an agent loads on demand	docs
awpack	First-party agent packs — the ones we build, versioned and installable on their own	docs
awm	A portable, scoped agent memory	docs
awnode	A lightweight local gateway — bridges your apps to the AI backends you chose	docs
awrun	A priority-aware queue and dispatcher for agentic runs and ad-hoc CI builds. It also judges whether the runner pool is big enough for the queue it is draining, and can ask a host to grow it -- reserving capacity is zero-sum, so a saturated pool needs more of it, not a different share of it	docs
awgraph	A semantic code graph for agents — AST + tree-sitter, call graphs	docs
awgit	Semantic version control on top of git — edit-ops and leases	docs
awdelphi	Anonymous multi-round expert panels — a converged answer with a trace	docs
awtoll	What every tool call costs you in context, measured from your own transcripts	docs
awseal	Sign an artifact so a stranger can verify it	docs
awshare	Publish an artifact and fetch it back verified	docs
awdit	An append-only audit trail whose gaps are DETECTABLE	docs
awbac	Role-based access control that fails closed and explains itself	docs
awiam	Who is this caller? A directory and session store that fails honestly	docs
awtunnel	Reach a service that has no public address	docs
awnest	Prove there is a human before you let them into the nest	docs
awnboard	A front gate you can put in front of anything, and hand someone the key to	docs
awnix	A Linux you can hand to an agent — immutable base, capabilities included	docs
awrecover	Labelled snapshots with an all-or-nothing restore	docs
awstorage	Every drive on every node, indexed, classified and diffed -- so you can see what you own before you delete it	docs
awrelay	Portable agent messaging — findings, alerts, coordination	docs
awask	Your agent asks you a question — and acts on your answer	docs
awmail	Give an agent an email address — send, and actually receive	docs
awnet	The agentic web — agents host a mesh, and agents join one	docs
awfind	A portable search client — query, results, ranking	docs
awbrowse	A portable browser client — navigate, console, network, DOM, screenshot	docs
awknowledge	How to run a coding agent so the result survives — the laws, with evidence	docs
gawbbonet	GobboNet campaigns with a real agent brain — scoped memory, graph recall	docs
aitherkvcache	Near-optimal KV cache quantization for LLM inference — sub-byte compression	docs
awrtifact	Deliberately chunk artifacts into GitHub release assets — the productized aitherkvcache mirror lane	docs
AitherZero	PowerShell 7+ automation framework — numbered, self-describing scripts	docs
AitherConnect	Browser extension — federated AI search, page context, and the Living OS overlay	docs
awreason	A portable reasoning client — sessions, phases, thoughts, and the chain that produced the answer	docs
awrecurse	Answer a question over a context far larger than the window — recursively, with the trace kept	docs
awprism	Turn a failure into ranked hypotheses — and say what would confirm each one	docs
awrepl	A REPL an agent can actually use — state that survives between turns	docs
awresearch	Ask a research question, get a cited report you can check	docs
awfocus	See, search and steer every Claude session from one command	docs
awgym	An ARC training gym — a game a world model can watch, and six roles that play through it	docs
awpredict	Predict what your environment does next, and how surprised you were	docs
awsh	Your terminal answers you -- type a question where a command would go	docs
awkno	The man page for the Aither World — every brick, stack and law, offline	docs
awembed (you are here)	Train an embedding model that knows your corpus, and prove it beats the big one	docs
awtax	Turn any tax PDF -- returns, W-2, 1099, statements, even scans -- into structured data you can check	docs
awsettings	Your agent's permissions and config, following you to the next machine	docs

<script src="aither-constellation.js"></script>