Why high: an exact fingerprint shared with 40 models that declare the same family, without a second signal of a different kind yet.
Relace: Relace Search is Relace's model. Its measured tokenizer sits with the Qwen group, which is evidence that it builds on the Qwen tokenizer. Tokenizer reuse is normal engineering, open vocabularies travel between labs, and this says nothing against Relace's authorship of the model itself.
| We checked | What we found | Where it came from |
|---|---|---|
| How it counts tokens | Counts every one of the 50 test strings exactly like 40 other models | tk_5410b38e |
| How its API is set up | A combination of settings no other listing in the catalogue uses | api_ea2ac21e |
| How much it can read and write | Reads up to 256,000 tokens at once, writes up to 128,000 | its own listing |
| Who served our requests | Relace | we measured it, 3 September 2026 |
Click a row to open the full comparison.
We sent Relace: Relace Search fifty short pieces of text and recorded what each one cost it in tokens. The bars below are those costs. Two models built on the same tokenizer produce the same bars; a model built on a different one produces a different set, which is what makes this a fingerprint.
Overhead subtracted: 8 prompt tokens.
The relace-search model uses 4-12 `view_file` and `grep` tools in parallel to explore a codebase and return relevant files to the user request. In contrast to RAG, relace-search performs agentic...
Supported parameters
max_tokensresponse_formatseedstoptemperaturetool_choicetoolstop_p
defaults: {"temperature":null,"top_p":null,"frequency_penalty":null}
