We shrank Kimi K3.1 Pro by 35%, within a point of the original on MMLU.

Each of the eight cards holds one slice of the model. Whatever the model leaves free, the server gives to conversations: the text the model keeps in memory while it answers. This is one card for each build, all drawn to one scale.

Moonshot publishes Kimi K3.1 Pro already stored in 4 bits, so every percentage on this page is against a model that is already compressed, not a 16-bit one. All three builds ran on the same eight B300 cards, with the same serving software, and answered the same questions. A shorter bar is a smaller build.

We asked three things of each build. Does it still know what the original knows? MMLU is 14,042 multiple-choice questions across 57 subjects, and higher is better. Does it make things up? Each build wrote out the facts in 360 passages, and we counted the facts no passage supports, so lower is better. Can it still use tools? It made 600 calls at each of two thinking efforts, scored on calling the right tool the right way, and higher is better.

Before the run we set the pass mark for MMLU: no more than one point below the original, compared question by question. The chart shows how far each build is below the original, and how far below it could be at worst. Left of zero is below the original, and the dashed line is the pass mark.

On made-up facts and tool calls both builds are level with the original. Tap a figure for the counts behind it.

A test score can hide small changes in how a model writes. Next-word agreement is a finer check: each build reads the same 300 passages, and at each of 82,712 places we ask whether it would pick the same next word as the original. Higher is closer to the original. The second column measures how far its word choices drift from the original’s, where lower is closer.

Every card has the same memory. What the weights leave free, the server gives to the cache: the conversation the model holds while it answers. A bigger cache means more people, or longer conversations, on the same cards. The server was allowed 90% of each card for every build, and a longer bar holds more.

Choose TWX3 when matching the original matters most: it picks the original’s next word more often. Choose TWX2.5 when room matters most: it carries 170 GB less weight than TWX3 and has 24% more room in the cache. On MMLU the two score the same.

The original and both builds were served the same way, one after another on one machine, and asked the same questions. The settings for each test follow.

The machine. Eight B300 cards, with vLLM’s own image for Kimi K3.1 serving each build from its own weights, split across all eight, with the engine’s 8-bit cache and 90% of each card.

Size. The weights files, in decimal gigabytes.

MMLU. Five-shot, no sampling, scored by the answer letter the model rates most likely. The paired test counts the answers that differ: 301 right only for the original and 218 only for TWX2.5 (p = 0.0003); 267 only for the original and 187 only for TWX3 (p = 0.0002). Between the two builds, 290 and 287 (p = 0.93).

Made-up facts. Each build reads 360 passages from Tileward’s own document set and writes out the facts they state. Every fact is matched back to its passage, and one that no passage supports counts as made up. The test treats facts as independent.

Tool calls. Six cases, 20 trials each, run five times: 600 calls, once at low thinking effort and once at high. The cases are one call, two calls in one turn, nested arguments, a tool that must not be called, the right tool out of twenty, and a two-turn loop. Each call is scored on tool name, call count and argument shape, with vLLM’s Kimi K3 tool parser.

Agreement. Each build reads the same 300 passages (82,712 token positions), and we count where its top next token is the original’s.

The model. Kimi K3.1 Pro is Moonshot AI’s, released under the Kimi K3.1 Pro License.