Kanana-2-v-decide from Kakao scores text and images with cached candidate embeddings. Check its token limits, custom runtime and licence terms.
By Kakao · Weights upload: 6 Nov 2026
A text/image decision model built on Kanana-2-v-3b. It compares state/question embeddings with candidate embeddings and returns option probabilities. Candidate embeddings can be cached. The supplied HTTP server processes one request at a time.
Quick answers
What is Kanana-2-v-decide by Kakao?
Kanana-2-v-decide: A text/image decision model built on Kanana-2-v-3b.
3.67B: 3.18B language model and 0.49B vision encoder
Inputs
Korean or English text, JSON or chat; up to 10 images; no video
Context
20,480 tokens; overlong inputs are shortened by default, or rejected with --strict-inputs
Licence
Kanana Open License
Access
Public weights, custom Transformers code and /v1/systemone server
Claims and sources
Kakao reports 61.34 on public Decision Index text and 61.12 on vision, averaging 61.23 across those two scores. Its Korean benchmark score is 72.50 over 15,909 questions. These are maker results, not a combined ranking of all decision models. vendor reportedsource
Pinned BF16 Docker recipe for Kanana on one NVIDIA GB10. Published memory, HTTP latency and Korean-exam records include actual choice, noul and score responses. The maker measured up to 17.12 GiB additional unified system memory under standalone load. A final image rebuild did not repeat full GPU loading; earlier loader runs and live-server smoke results remain separate.
yangs1202 · ★ 0 · MIT recipe; Kanana Open License weights and NVIDIA image terms · Kanana-2-v-decide