AI Desk · Model Identity · Compare
OpenAI: GPT-5 Nano · Z.ai: GLM 5.3
Identity similarity: lowDivergent fingerprints · 18/50These models tokenise differently. Divergence on the high-information strings (CJK, emoji, digits) separates unrelated vocabularies within a handful of samples.
OpenAI: GPT-5 Nano recordZ.ai: GLM 5.3 recordOpen in the interactive bench
Tokenizer fingerprint
18 of 50 mutually clean strings agree| Test string | OpenAI: GPT-5 Nano | Z.ai: GLM 5.3 | Verdict |
|---|---|---|---|
| en-prose | 14 | 14 | match |
| en-long | 17 | 16 | differs |
| spaces-20 | 3 | 3 | match |
| spaces-60 | 3 | 3 | match |
| tabs-20 | 3 | 3 | match |
| newlines-20 | 4 | 4 | match |
| mixed-ws | 6 | 5 | differs |
| digits-9 | 3 | 5 | differs |
| digits-12 | 4 | 7 | differs |
| digits-30 | 10 | 10 | match |
| digits-sep | 7 | 9 | differs |
| float-long | 9 | 14 | differs |
| zh-common | 12 | 9 | differs |
| zh-long | 17 | 10 | differs |
| zh-rare | 20 | 18 | differs |
| ja-kana | 10 | 10 | match |
| ja-kanji | 13 | 14 | differs |
| ko | 11 | 17 | differs |
| ru | 11 | 13 | differs |
| ar | 11 | 13 | differs |
| he | 10 | 24 | differs |
| hi | 11 | 27 | differs |
| th | 11 | 24 | differs |
| el | 14 | 14 | match |
| emoji-basic | 8 | 5 | differs |
| emoji-skin | 13 | 10 | differs |
| emoji-zwj-family | 11 | 7 | differs |
| emoji-zwj-x3 | 33 | 21 | differs |
| emoji-flags | 16 | 24 | differs |
| emoji-prof | 20 | 15 | differs |
| math | 31 | 27 | differs |
| boxdraw | 27 | 28 | differs |
| combining | 11 | 10 | differs |
| cjk-ext-b | 13 | 13 | match |
| surrogates | 40 | 40 | match |
| zalgo | 35 | 22 | differs |
| rtl-mix | 5 | 8 | differs |
| py-code | 25 | 25 | match |
| py-indent | 15 | 15 | match |
| json | 20 | 21 | differs |
| html | 16 | 16 | match |
| regex | 44 | 44 | match |
| camel | 6 | 5 | differs |
| snake | 6 | 6 | match |
| rare-word-x5 | 31 | 31 | match |
| repeat-tok | 22 | 22 | match |
| base64 | 34 | 35 | differs |
| hex | 11 | 14 | differs |
| url | 17 | 17 | match |
| uuid | 27 | 28 | differs |
Values are the marginal prompt-token cost of each string with the model's fixed overhead subtracted. An x marks a row excluded as corrupt for that model.
API surface
The declared serving contracts differ| Field | OpenAI: GPT-5 Nano | Z.ai: GLM 5.3 | Verdict |
|---|---|---|---|
| Context length | 400000 | 1048576 | × differs |
| Max output tokens | 128000 | 131072 | × differs |
| Supported parameters | include_reasoning, max_completion_tokens, max_tokens, reasoning, reasoning_effort, response_format, seed, structured_outputs, tool_choice, tools | include_reasoning, max_tokens, reasoning, reasoning_effort, response_format, temperature, tool_choice, tools, top_k, top_p | × differs |
| Defaults | {"temperature":null,"top_p":null,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null} | {"temperature":1,"top_p":0.95} | × differs |
| Reasoning contract | {"mandatory":true,"supported_efforts":["high","medium","low","minimal"],"default_effort":"medium"} | {"mandatory":true,"default_enabled":true,"supported_efforts":["max","high","low"],"default_effort":"max"} | × differs |
| Modalities | text+image+file->text | text->text | × differs |
Method: tokenizer fingerprinting, v1, 50 strings. Raw data on the data page. This page is static and dated; the interactive bench compares any two of the 422 catalogue models.