Evidence & limits

Evidence

• A live network. Pearl's integer whitepaper states that the chain became live when the node code was made public ². The repository supports mainnet and test networks, and has published release v1.2.1 ⁴.
• A serving benchmark. Pearl reports a benchmark on four H200 GPUs comparing Llama 3.3 70B with its "two-for-one" Pearl-certified variant, which re-implements a layer with a new quantisation mechanism ²:
  • the variant reached 17,206.26 tokens per second with pipeline parallelism, and 18,291.66 with four-way data parallelism;
  • the original bf16 model reached at most 15,269.81 tokens per second, and ran out of memory with four-way data parallelism;
  • MMLU scores were close: 0.8180 to 0.8198 for the variant against 0.8193 to 0.8198 ².
• Kernel overhead. In a separate benchmark against stock serving engines, Pearl reports 5.08% end-to-end overhead for Llama 70B with four-way data parallelism on four H200 GPUs, and 3.9% for DeepSeek V3.2 with eight-way data parallelism plus expert parallelism on eight H200 GPUs ³. These are developer-reported measurements of its integer proof-of-useful-work kernel.
• The underlying theory. The construction behind Pearl proves a multiplicative overhead of 1 + o(1) over naive matrix multiplication ⁵. Pearl's FP8 specification states a "1 + o(1) factor" overhead, without a measured percentage ¹.
• Independent measurement. Basu measured Pearl's mainnet, running the integer scheme, in May 2026, counting 8,012 online workers among the top 15 miners on the AlphaPool mining pool, which Basu estimates to hold about 21% of network hashrate. String analysis suggests that the dominant mining software, from a third party, contains no inference code and generates matrices from random seeds. Random matrices passed verification in the study's tests, and a mining pool accepted shares mined with them ⁶.

Limitations

Known speedups. Pearl documents known mining speedups and the checks against them ¹. It treats faster honest kernels or hardware as outside the threat model ¹.

Hardness assumption. Security rests on an informal hardness assumption specific to quantised, noised matrices ¹.

Usefulness is not verified. Verification checks the multiplication, not where the matrices came from ⁶.

Benchmark caveats. The benchmark compares a Pearl-certified model with the original model, and reports no run of the same certified model without mining ².

Fit to verification. Pearl's proofs show that work was performed, not that no other work ran. Attestable, which proposes proof-of-work accounting for AI agreements, notes that "a proof of some computation is not a proof of all computation", and that bounding spare capacity needs a credible estimate of the compute available ⁸.