fix(rank): 看图把关候选编号统一(3B 视觉模型 1/5 → 4/5)+ 可选本机 Ollama 视觉模型 - #22
Merged
Merged
Conversation
jnMetaCode
force-pushed
the
fix/rank-index-base
branch
from
September 22, 2026 11:27
7ea9d56 to
6998f4a
Compare
提示词里"候选 0:[图片1]"两套编号打架,本机 3B 视觉模型 3 选 1 只对 1/5(第 1 条永远算 0 分);统一后 4/5。 ollama 进看图把关的供应商列表,型号只列本机能看图的(需引擎 ≥ 0.19.3 把图真发出去;旧引擎验证会如实报看不了图)。
jnMetaCode
force-pushed
the
fix/rank-index-base
branch
from
September 22, 2026 13:50
6998f4a to
8f84ad9
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
两件事
i,第 1 条候选永远被解析成 0 分。现在候选从 1 起编号,解析兼容仍按 0 起数的模型(出现 i=0 就按 0 起)。同一模型同一批 12 张缓存图:1/5 → 4/5。单测 + 变异检查。feat/ollama-vision,未推):以前剥掉图片,现在走images字段,并把图片算进 num_ctx(真机 12 张缩略图只按文本估 4096 → Ollama 400)。旧引擎下"验证并开启"真发一张红图,会如实报看不了图,不会假通过。真机(qwen2.5vl:3b,AO 补丁临时放进 node_modules 验的,验完还原)
openshorts run --vision-provider ollama --only s1:一镜从"本机出图 56 秒"变成"本机看图选中素材 44 秒"ollama pull qwen2.5vl:7b(7B 未验,要再下 6 GB)全量测试通过、构建通过。