[PR #2389] [CLOSED] Allows settings of rope_freq_base and rope_freq_scale again in modelfile #73168

Closed
opened 2026-05-05 04:51:50 -05:00 by GiteaMirror · 0 comments
Owner

📋 Pull Request Information

Original PR: https://github.com/ollama/ollama/pull/2389
Author: @jukofyork
Created: 2/7/2024
Status: Closed

Base: mainHead: jukofyork-rope-settings


📝 Commits (10+)

📊 Changes

7 files changed (+78 additions, -8 deletions)

View changed files

📝 api/types.go (+7 -2)
📝 docs/api.md (+2 -0)
📝 docs/modelfile.md (+3 -0)
📝 llm/dyn_ext_server.go (+8 -2)
📝 llm/ext_server/ext_server.cpp (+53 -1)
📝 llm/ext_server/ext_server.h (+5 -1)
📝 llm/llm.go (+0 -2)

📄 Description

This adds back the ability to set rope_freq_base and rope_freq_scale again in the model file.

If the values aren't set then defaults to passing 0.0f for both to the llama.cpp server which in turn gets the values from the GGUF file itself.


This also includes the code for the PR that allows split_mode and tensor_split to be set from the modelfile (Github won't let me make a second fork and I'm too dumb to work out how to split off just the changes for the rope_freq_base and rope_freq_scale - sorry).


🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.

## 📋 Pull Request Information **Original PR:** https://github.com/ollama/ollama/pull/2389 **Author:** [@jukofyork](https://github.com/jukofyork) **Created:** 2/7/2024 **Status:** ❌ Closed **Base:** `main` ← **Head:** `jukofyork-rope-settings` --- ### 📝 Commits (10+) - [`1e14d5c`](https://github.com/ollama/ollama/commit/1e14d5cd0eeb4fb78b835fe73aede7a84ef8359f) Update types.go - [`46db6c4`](https://github.com/ollama/ollama/commit/46db6c46d162fccc94124aef4aa73fbba63db8f1) Update api.md - [`ee644b2`](https://github.com/ollama/ollama/commit/ee644b22aa26ccff7cc32cc3efdc3ec8128b55ea) Update modelfile.md - [`c4d8f7a`](https://github.com/ollama/ollama/commit/c4d8f7ae7f94e522404ef1e32bb7706eef0465a5) Update dyn_ext_server.go - [`311ffda`](https://github.com/ollama/ollama/commit/311ffdaa877684ad6158291f96e3625a2d343f2b) Update types.go - [`5042b82`](https://github.com/ollama/ollama/commit/5042b82141d41846478aa04dec4b7a15588f06bd) Update ext_server.h - [`b9cfce9`](https://github.com/ollama/ollama/commit/b9cfce970214299e9eb2799516a42c95b74b0a73) Update ext_server.cpp - [`ab55ae1`](https://github.com/ollama/ollama/commit/ab55ae1a590a109320f2e4a0f053e55c63614e76) Update dyn_ext_server.go - [`a03342f`](https://github.com/ollama/ollama/commit/a03342f734153134fc4e0142bd9024d44278caf0) Update api.md - [`f31b657`](https://github.com/ollama/ollama/commit/f31b6570272b41544ab85f3be570e353fa61dbc1) Update types.go ### 📊 Changes **7 files changed** (+78 additions, -8 deletions) <details> <summary>View changed files</summary> 📝 `api/types.go` (+7 -2) 📝 `docs/api.md` (+2 -0) 📝 `docs/modelfile.md` (+3 -0) 📝 `llm/dyn_ext_server.go` (+8 -2) 📝 `llm/ext_server/ext_server.cpp` (+53 -1) 📝 `llm/ext_server/ext_server.h` (+5 -1) 📝 `llm/llm.go` (+0 -2) </details> ### 📄 Description This adds back the ability to set `rope_freq_base` and `rope_freq_scale` again in the model file. If the values aren't set then defaults to passing `0.0f` for both to the `llama.cpp` server which in turn gets the values from the GGUF file itself. --- This also includes the code for the PR that allows `split_mode` and `tensor_split` to be set from the modelfile (Github won't let me make a second fork and I'm too dumb to work out how to split off just the changes for the `rope_freq_base` and `rope_freq_scale` - sorry). --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>
GiteaMirror added the pull-request label 2026-05-05 04:51:51 -05:00
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#73168