Why high: an exact fingerprint shared with 5 models.
IBM: Granite 4.0 Micro is IBM's model. Its measured tokenizer sits with the OpenAI cl100k group, which is evidence that it builds on the OpenAI cl100k tokenizer. Tokenizer reuse is normal engineering, open vocabularies travel between labs, and this says nothing against IBM's authorship of the model itself.
cl100k_base is an open vocabulary table anyone can build on. Sharing it says which dictionary the model reads with, and nothing about who made the model.
Granite 4.0's endpoint counts match OpenAI's open cl100k_base encoding on all 50 strings, and its repository declares the matching 100,352-entry vocabulary. Microsoft's Phi-4 measures identically, which is why two rival labs share one signature group: both adopted the tokenizer OpenAI released as open source with tiktoken.
Granite 4.0 H Micro repository (config.json: vocab_size 100352)OpenAI tiktoken (cl100k_base, open source)
| We checked | What we found | Where it came from |
|---|---|---|
| How it counts tokens | Counts every one of the 50 test strings exactly like 5 other models | tk_ba918336 |
| How its API is set up | A combination of settings no other listing in the catalogue uses | api_10a7f490 |
| How much it can read and write | Reads up to 131,000 tokens at onceOpenRouter's listing shows 117,900 output tokens, which is 90% of the context window, the figure it carries where the provider declares no maximum. | its own listing |
| Who served our requests | Cloudflare | we measured it, 21 August 2026 |
Click a row to open the full comparison.
We sent IBM: Granite 4.0 Micro 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: 30 prompt tokens.
Granite-4.0-H-Micro is a 3B parameter from the Granite 4 family of models. These models are the latest in a series of models released by IBM. They are fine-tuned for long...
Supported parameters
frequency_penaltylogit_biaslogprobsmax_tokensmin_ppresence_penaltyrepetition_penaltyresponse_formatseedstoptemperaturetop_ktop_logprobstop_p
defaults: {"temperature":null,"top_p":null,"frequency_penalty":null}
