sleepy/llama.cpp
1
0
Fork 0
Code Issues 16 Pull Requests Actions 133 Packages Projects Releases Wiki Activity

model : support Step3.5-Flash (#19283)

Browse Source 
* Support Step3.5-Flash

* fix: norm.weight + 1 (HF zero_centered=true)

* step35: simplify GGUF conversion + drop redundant rope KVs

* Address review feedback

* rename limits -> clamp

* Apply suggestions from code review

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>

* Apply suggestion from @CISC

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>

* rename swiglu limits -> swiglu clamp in LLM_KV

* avoid CI fail

* Apply suggestions from code review

* Apply suggestions from code review

* disabled KV shifting for LLM_ARCH_STEP35

* Apply suggestions from code review

* mistakenly removed cmath

* add model size && apply missed suggestion

* assert partial_rotary_factors

* fix CI errors:

* load freq_base_swa

---------

Co-authored-by: lvyichen <lvyichen@stepfun.com>
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
This commit is contained in:
forforever73
2026-02-07 04:06:14 +08:00
committed by GitHub
parent 3228e77287
commit b83111815e
15 changed files with 576 additions and 38 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/llama-kv-cache.cpp
+4
View File
@@ -974,6 +974,10 @@ void llama_kv_cache::apply_ubatch(const slot_info & sinfo, const llama_ubatch &
}
bool llama_kv_cache::get_can_shift() const {
// Step35 uses per-layer RoPE dims; K-shift assumes a single global n_rot.
if (model.arch == LLM_ARCH_STEP35) {
return false;
}
return true;
}