Incompressible Knowledge Probes: Estimating Black-Box LLM Parameter Counts via Factual Capacity · HackerPapers discussions

Incompressible Knowledge Probes: Estimating Black-Box LLM Parameter Counts via Factual Capacity — mentioned by WarmWash

Wed, 02 Sep 2026 23:43:17 GMT

It's not totally a mystery https://arxiv.org/html/2604.24827v1 The short of it is by using hard facts knowledge that is difficult to compress, and then quizzing models on these facts and calibrating against a bunch of open models, you can kind of feel out the size of closed models.

Paper and discussion history. A mention is not an endorsement.

Incompressible Knowledge Probes: Estimating Black-Box LLM Parameter Counts via Factual Capacity — mentioned by thoughtpeddler

Thu, 06 Aug 2026 22:09:20 GMT

Oh ya, this is like the approach from the Incompressible Knowledge Probes [0] paper - smart! [0] Incompressible Knowledge Probes: Estimating Black-Box LLM Parameter Counts via Factual Capacity [https://arxiv.org/abs/2604.24827]

Paper and discussion history. A mention is not an endorsement.

Incompressible Knowledge Probes: Estimating Black-Box LLM Parameter Counts via Factual Capacity — mentioned by Chyzwar

Tue, 26 May 2026 18:56:17 GMT

https://arxiv.org/abs/2604.24827 From this paper

Paper and discussion history. A mention is not an endorsement.

Incompressible Knowledge Probes: Estimating Black-Box LLM Parameter Counts via Factual Capacity — mentioned by WarmWash

Wed, 20 May 2026 14:30:17 GMT

Opus is estimated to be around 4T parameters, and 5.5 around 9T. [1] And while 3.5 at least qualifies to be in the same neighborhood, which is stunning if these numbers are all true, it may be that closing that last ~10% difference needs 50x more parameters. [1]https://arxiv.org/pdf/2604.24827

Paper and discussion history. A mention is not an endorsement.

Incompressible Knowledge Probes: Estimating Black-Box LLM Parameter Counts via Factual Capacity — mentioned by schmorptron

Sat, 02 May 2026 10:56:52 GMT

Associated paper: https://arxiv.org/html/2604.24827v1

Paper and discussion history. A mention is not an endorsement.

Incompressible Knowledge Probes: Estimating Black-Box LLM Parameter Counts via Factual Capacity — HN submission

Thu, 30 Apr 2026 09:38:42 GMT

Estimating Black-Box LLM Parameter Counts via Factual Capacity

3 points; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.