SmolLM2-135M compiled leg: DSL → StableHLO → IREE showcase (gemma-iree precedent)
维护者通常 1 天内回复
还没有人认领这个 Issue。
评估
- 难度
- 5/5
- 预计耗时
- 一周以上
- 新手友好度
- 25/100
- Issue 类型
- 功能
- 描述清晰度
- 基本清楚
- 活跃度
- 冷清
- 技术栈
- kotlin
调研方向
Read llm-runtime/gemma-iree/src/nativeMain/IreeRuntime.kt, GemmaDecoder.kt, and GemmaKvDecoder.kt, then inspect the host-side :llm-inference:gemma StableHLO export. The issue asks for a SmolLM2 export and compiled decode path integrated with the #272 cross-target reproducer; done means eager and compiled results can be compared on the listed targets. This spans model export, runtime architecture, and multiple platforms.
由索引模型根据 Issue 内容生成。
描述
Split out from #272 — the eager-path scope of that issue is done (the
SmolLm2InferenceSpike.kt commonTest reproducer, AndroidSmolLm2E2eTest.kt,
and the feature/smoll disposition are all resolved). This issue tracks the
one item that's still open: the compiled path.
Ask
Add the SmolLM2-135M-Instruct compiled leg — DSL → nn-DAG → StableHLO → IREE
— following the :llm-runtime:gemma-iree precedent:
- A
gemma3(orsmollm2) → StableHLO export on the host side, matching
:llm-inference:gemma's existing gemma3 → StableHLO export that produces
the.vmfb+.irpapair. - A
:llm-runtime:smollm2-iree-shaped module (or extendgemma-ireeif the
architecture is close enough) with the on-device decode loop —
IreeRuntime+ a*Decoder, mirroringGemmaDecoder/GemmaKvDecoderin
llm-runtime/gemma-iree/src/nativeMain. - Wire it into the same cross-target reproducer harness from #272 so eager
vs. compiled numbers are directly comparable per target (JVM, Android,
iOS simulator/device, linuxX64/Arm64, macosArm64).
Why this model
SmolLM2-135M-Instruct (138 MiB, GGUF Q8_0) is already the standard
cross-target reproducer per #272 — small enough for CI, phones and
simulators, llama-architecture. Using the same checkpoint for the compiled
leg gives an apples-to-apples eager-vs-compiled comparison per target instead
of introducing a second model just for this path.
Context
- Field report that motivated the reproducer effort: SKaiNET-developers/SKaiNET's mobile
field report (context in #272). gemma-ireeprecedent:llm-runtime/gemma-iree(IreeRuntime.kt,
GemmaDecoder.kt,GemmaKvDecoder.ktinsrc/nativeMain), paired with the
host-sidegemma3→ StableHLO export in:llm-inference:gemma.
- 主要语言
- Kotlin
- 星标
- 8
- 派生
- 3
- 平均合并
- 1 天 16 小时
- 30 天内合并 PR
- 48
环境准备
这个项目没有提供开发容器、Dockerfile 或贡献指南，环境需要你自己搭建：先看它的 README，通用步骤见我们的新手贡献指南。
从这里开始
- 先读完整个 Issue，再读项目的贡献指南。
- 在 Issue 下留言说明你要接手 —— 这能避免两个人做同样的事。
- Fork 仓库，在一个分支上完成修改。
- 提交 Pull Request，并在描述里引用这个 Issue 编号。
SKaiNET-developers/SKaiNET-transformers 的其他 Issue
-
documentation good first issue size:xs skill:docs sub-issue
难度 2/5 1-3 小时 新手友好度 78/100
SKaiNET-developers/SKaiNET-transformers#478 · 1 条评论 ·
维护者通常 1 天内回复
-
enhancement skill:kotlin-core
难度 4/5 3-5 天 新手友好度 48/100
SKaiNET-developers/SKaiNET-transformers#502 ·
维护者通常 1 天内回复
-
bug
难度 4/5 3-5 天 新手友好度 48/100
SKaiNET-developers/SKaiNET-transformers#483 ·
维护者通常 1 天内回复
-
coding good first issue size:s skill:kotlin-core sub-issue
难度 4/5 1-2 天 新手友好度 64/100
SKaiNET-developers/SKaiNET-transformers#477 · 1 条评论 ·
维护者通常 1 天内回复
-
coding good first issue size:s skill:numerics sub-issue
难度 4/5 3-5 天 新手友好度 55/100
SKaiNET-developers/SKaiNET-transformers#476 · 1 条评论 ·
维护者通常 1 天内回复
查看 SKaiNET-developers/SKaiNET-transformers 的全部 Issue
相似的 Issue
-
难度 2/5 1-3 小时 新手友好度 82/100
维护者通常 1 天内回复
-
enhancement impact:U1 P3 reach:R2 size:S
难度 2/5 1-3 小时 新手友好度 78/100
-
enhancement triaged
难度 1/5 1 小时以内 新手友好度 75/100
维护者通常 1 天内回复
-
难度 2/5 1-3 小时 新手友好度 73/100
ankidroid/Anki-Android#22677 ·
维护者通常 1 天内回复
-
Type Pick Request
难度 2/5 1-3 小时 新手友好度 68/100
reactwg/react-native-releases#1456 · 1 条评论 ·