README.md

metadata

license: cc0-1.0
language:
  - en
pretty_name: Ballast T0
size_categories:
  - 100M<n<1B
tags:
  - knowledge-base
  - wikidata
  - rag
  - grounding
  - knowledge-graph
  - hallucination
configs:
  - config_name: entities
    data_files: entities/tier=T0/lang=en/rank_bucket=*/*.parquet
  - config_name: triples
    data_files: triples/tier=T0/lang=en/rank_bucket=*/*.parquet
  - config_name: properties
    data_files: properties.parquet

⚓ Ballast T0

We measured how much of a bigger model's factual advantage is just memorized trivia, and whether you can buy that back with a file instead of with parameters. You can, and it's 40–100× cheaper per byte.

This is that file: the reference ballast — a versioned, rank-quantized knowledge artifact you pair with any language model. 25.4M entities, 197.4M cleaned Wikidata triples, 1.51 GB total, published in 8 nested rank buckets: pick your knowledge level like you pick a GGUF quant.

Measured effect (50,147 linked probes, logprob choice scoring):

Gemma-4	raw	+ full ballast	hallucination
E2B	0.608	0.868	0.242 → 0.074
E4B	0.662	0.910	0.197 → 0.042
12B	0.683	0.910	0.207 → 0.049

Qwen3.5	raw	+ full ballast	hallucination
0.8B	0.324	0.784	0.601 → 0.114
2B	0.363	0.770	0.552 → 0.137
4B	0.434	0.831	0.474 → 0.095
9B	0.542	0.819	0.329 → 0.113

The pattern replicates across two unrelated families: raw floors spread, ballasted ceilings compress — and the ballasted Qwen 4B beats the ballasted 9B outright. With perfect entity resolution, +180 MB of ballast lifts Gemma E2B past the 12B's raw factual accuracy; with the shipped non-generative linker (which realizes about two thirds of that ideal gain), the crossing needs ~470 MB. Either way the parameter route to the same gain costs ~19 GB — a ~40× byte disadvantage.

Two more measured effects worth knowing:

• Ballasted accuracy is a quantization-damage diagnostic. 4-bit (nf4) leaves Gemma E2B and 12B nearly intact but collapses E4B's ballasted ceiling from 0.910 to 0.650 — when quantization breaks a model's ability to read evidence, no corpus buys it back, and the grounded score is what reveals it.
• This fixes answerable questions; it does not teach abstention. On 2-hop composition probes, grounding cuts hallucination 3–20×. On questions with no true answer (false premises, never-recorded facts), fabrication rises with corpus coverage — evidence reads as license to answer.

Full thesis, methodology, and caveats: github.com/OpenBallast/ballast.

Layout & quantization levels

Hive-partitioned Parquet; the partition key is the quantization axis. A level Lk = buckets 0..k — download fewer bucket directories to hold a smaller corpus:

entities/tier=T0/lang=en/rank_bucket={0..7}/*.parquet
triples/tier=T0/lang=en/rank_bucket={0..7}/*.parquet
properties.parquet    # 13,704 property labels, shared by every level
manifest.json         # counts, sizes, hashes, build provenance

level	cumulative	entities
L0	36 MB	top 0.5%
L1	62 MB	top 1%
L2	107 MB	top 2%
L3	179 MB	top 4%
L4	288 MB	top 8%
L5	466 MB	top 16%
L6	756 MB	top 32%
L7	1,507 MB	all 25.4M

Ranking = composite of log(sitelinks) + log(claim count). Buckets nest: L3 is a byte prefix of L7. Truncation degrades honestly — an entity outside the level has no evidence; a triple whose object falls outside the level is dropped, never rendered as a bare Q-id.

Schemas

entities: qid, label, description, aliases[], sitelinks, n_claims (+ rank_bucket from path) triples: qid, pid, value_type (entity|time|quantity|string|coord), value (+ subject's rank_bucket)

Excluded at publish: external identifiers, media references, URLs — bytes that ground nothing a language model can use.

Use it

DuckDB (the layout is built for it):

SELECT t.qid, p.label AS prop, t.value
FROM read_parquet('ballast-t0/triples/**/*.parquet', hive_partitioning=true) t
JOIN read_parquet('ballast-t0/properties.parquet') p USING (pid)
WHERE t.rank_bucket <= 3 AND t.qid = 'Q42';

Evidence rendering (the exact format the published numbers were measured under):

Facts about Douglas Adams:
- place of birth: Cambridge
- occupation: science fiction writer
...

Run it locally (serving format)

serving/sqlite/ carries the same corpus as zstd-compressed per-bucket SQLite — the format consumed by the ballast CLI:

uvx openballast pull --level 3     # 265 MB download -> ~1 GB on disk
uvx openballast serve              # OpenAI grounding proxy :11435 + MCP :11436

level	download	on disk		level	download	on disk
L0	52 MB	0.2 GB		L4	427 MB	1.6 GB
L1	92 MB	0.35 GB		L5	691 MB	2.6 GB
L2	159 MB	0.6 GB		L6	1.1 GB	4.2 GB
L3	265 MB	1.0 GB		L7	2.2 GB	9.2 GB

Each bucket_k.sqlite holds that bucket's entities, names (normalized label/alias index), and triples; levels attach buckets 0..k, so upgrading a level downloads only the new buckets. Truncation semantics identical to the parquet artifact. Parquet remains the canonical/build format; the sqlite tree is a derived serving artifact.

Live demo endpoint

Levels L0–L5 are served free at https://mcp.openballast.org (MCP over streamable HTTP + plain GET) — running entirely on a $0/month free-tier stack as an existence proof. Demo-grade: no auth, no SLA. Reference: docs/mcp.md.

curl "https://mcp.openballast.org/lookup?question=Where+was+Douglas+Adams+born%3F&level=5"

Provenance & license

Built from the Wikidata JSON dump (2026-07 snapshot; parameters and hashes in manifest.json). mul language code honored with English fallback; Wikimedia-internal pages filtered by instance of. Data: CC0 (Wikidata contributors). Companion eval sets: OpenBallast/ballast-evalsets.