Report · resources

KTS-1 standard

Qwen3.8-27B · Mirai S trellis with Q8_0 MTP variant · Mirai S CUDA dev

Measured results / KTS-1 standard / Run 0388c0a8d393422585997e9e

Measured 2026-10-10 20:27 UTC · Record date 2026-10-10

Failed.

Answer checks

No answer-quality suite was graded for this run.

Method

KTS-1. Greedy deterministic requests through a verified Keep release. Three cold repeats per grid cell, five when spread exceeds ten percent; shared warm results are separate.

Limits

Labelled lab evidence for the recorded setup and standard. Completion is coverage, not customer or general answer-quality qualification.

What did not work

Standard check failed; the retained internal record contains the failure reason.

What ran

Exact identity

Original publisher model: Qwen3.8-27B
Publisher revision: not recorded
Variant · format · precision: Mirai S trellis with Q8_0 MTP variant · Mirai S trellis with Q8_0 MTP
Weights SHA-256: not recorded
Engine · version · feature: Mirai S CUDA · 0.2.0-dev-vector-disabled · plain
Hardware: NVIDIA GeForce RTX 3060, 12 GB VRAM
Harness / workload: not recorded
Workload suite · version: not recorded
Harness SHA-256: not recorded
Fixture SHA-256: not recorded

Workload shape

Full reproduction pack available to customers and evaluators on request

All recorded numbers

Metric definitions, scope and statistics

All aggregate numeric result fields in this run
Metric	Value	Scope	Statistic	Sample count	Basis / source field	Run
cold decode tps	39.35 tok/s	per person	single	not recorded	standard_result.cold_decode_tps	Run 0388c0a8 →

Data and citation

Download this run · OLIR JSON All numbers · CSV OLIR schema

Skilak Keep evidence, run 0388c0a8d393422585997e9e, measured 2026-10-10T20:27:37.404250+00:00.
https://skilak.ai/keep/resources/evidence/r/0388c0a8d393422585997e9e

Ready to scope it?

Start with one workload.

Tell us who uses it, what it can read, and what must stay inside. We will map the build.

Request a pilot