Why high: an exact fingerprint shared with 4 models.
MoonshotAI: Kimi K3 is Moonshot AI's model, and its measured tokenizer sits in Moonshot AI's own signature group alongside its sibling models. Measured models from Fireworks carry the same vocabulary, which reads as reuse of Moonshot AI's tokenizer line and questions nobody's authorship; they are listed under "Shares this fingerprint" below.
Re-measured on 24 September 2026 on a single provider, Fireworks, as the control for Fireworks' Ember-1. The 21 August sweep read this model while OpenRouter spread the calls across seven providers, and eight of its rows came back exactly 3 tokens low, the difference between two providers' fixed overheads, which left it without a family. On one provider all 50 strings are clean and match Kimi K2.6, K2.7 Code and K2 Thinking exactly: Moonshot AI kept one vocabulary from Kimi K2 to K3. Raw measurements are in kept with the desk's research files.
| We checked | What we found | Where it came from |
|---|---|---|
| How it counts tokens | Counts every one of the 50 test strings exactly like 4 other models | tk_ea6068bf |
| How its API is set up | Set up identically to 1 other listing, field for field | api_afde644c |
| How much it can read and write | Reads up to 1,048,576 tokens at onceOpenRouter's listing shows 943,718 output tokens, which is 90% of the context window, the figure it carries where the provider declares no maximum. | its own listing |
| Whether it thinks before answering | Reasons when asked to, at max, high, low, max by default | its own listing |
| Who served our requests | InferenceNet | we measured it, 5 October 2026 |
| Measured twice | The 5 October 2026 reading agreed with the 24 September 2026 reading on every one of the 50 strings clean in both: the tokenizer behind this name held. | 50/50 agree |
Click a row to open the full comparison.
We sent MoonshotAI: Kimi K3 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: 85 prompt tokens.
Kimi K3 is a 2.8T parameter open-weight multimodal reasoning model from Moonshot AI. It is suited for complex coding, knowledge work, and long-horizon agentic workflows, and is particularly strong at...
Supported parameters
frequency_penaltyinclude_reasoninglogit_biaslogprobsmax_tokensmin_ppresence_penaltyreasoningreasoning_effortrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_p
defaults: {"temperature":null,"top_p":0.95,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null}
