github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 16:40:08 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[PR #12105] Add support for the 'ignore_eos'. #60405

New Issue
Open
opened 2026-04-29 15:21:35 -05:00 by GiteaMirror · 0 comments
GiteaMirror commented 2026-04-29 15:21:35 -05:00
Owner
Copy Link

📋 Pull Request Information

Original PR: https://github.com/ollama/ollama/pull/12105
Author: @DoubleRedX
Created: 8/28/2025
Status: 🔄 Open

Base: main ← Head: main

📝 Commits (2)

  • a2d5e2b Add support for the 'ignore eos' parameter, which can be used in conjunction with 'num_predict' to specify the number of output tokens of the model.
  • 55b6b0d Modify test about openai.

📊 Changes

9 files changed (+29 additions, -1 deletions)

View changed files

📝 api/types.go (+2 -0)
📝 llama/llama.cpp/common/sampling.cpp (+4 -0)
📝 llama/llama.cpp/common/sampling.h (+2 -0)
📝 llama/llama.go (+6 -0)
📝 llama/sampling_ext.cpp (+5 -0)
📝 llama/sampling_ext.h (+2 -0)
📝 openai/openai.go (+3 -0)
📝 openai/openai_test.go (+3 -0)
📝 runner/llamarunner/runner.go (+2 -1)

📄 Description

Add support for the 'ignore eos' parameter, which can be used in conjunction with 'num_predict' to specify the number of output tokens of the model.

When using tools to conduct performance tests on Ollama's openai-compatible API, it is necessary to specify the size of the output token. However, in Ollama, the num predict parameter can only limit the upper bound of the number of tokens generated by the model. Therefore, the 'ignore eos' parameter is needed to enable the model to generate tokens infinitely to achieve the purpose of generating a specified number of tokens.

🔄 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/12105 **Author:** [@DoubleRedX](https://github.com/DoubleRedX) **Created:** 8/28/2025 **Status:** 🔄 Open **Base:** `main` ← **Head:** `main` --- ### 📝 Commits (2) - [`a2d5e2b`](https://github.com/ollama/ollama/commit/a2d5e2bd70ffe9f47225b70dd12d19821b0c3ecc) Add support for the 'ignore eos' parameter, which can be used in conjunction with 'num_predict' to specify the number of output tokens of the model. - [`55b6b0d`](https://github.com/ollama/ollama/commit/55b6b0d0d6278ca63b80942a5799a2f689a4c1e5) Modify test about openai. ### 📊 Changes **9 files changed** (+29 additions, -1 deletions) <details> <summary>View changed files</summary> 📝 `api/types.go` (+2 -0) 📝 `llama/llama.cpp/common/sampling.cpp` (+4 -0) 📝 `llama/llama.cpp/common/sampling.h` (+2 -0) 📝 `llama/llama.go` (+6 -0) 📝 `llama/sampling_ext.cpp` (+5 -0) 📝 `llama/sampling_ext.h` (+2 -0) 📝 `openai/openai.go` (+3 -0) 📝 `openai/openai_test.go` (+3 -0) 📝 `runner/llamarunner/runner.go` (+2 -1) </details> ### 📄 Description ### Add support for the 'ignore eos' parameter, which can be used in conjunction with 'num_predict' to specify the number of output tokens of the model. When using tools to conduct performance tests on Ollama's openai-compatible API, it is necessary to specify the size of the output token. However, in Ollama, the num predict parameter can only limit the upper bound of the number of tokens generated by the model. Therefore, the 'ignore eos' parameter is needed to enable the model to generate tokens infinitely to achieve the purpose of generating a specified number of tokens. --- <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 2026-04-29 15:21:35 -05:00
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
amd
api
app
bug
build
cli
cloud
compatibility
context-length
create
docker
documentation
embeddings
feature request
feedback wanted
good first issue
gpt-oss
gpu
harmony
help wanted
image
install
intel
js
launch
linux
macos
memory
mlx
model
needs more info
networking
nvidia
ollama.com
performance
pull-request
Mirrored from GitHub Pull Request
 python
question
registry
rendering
thinking
tools
top
vulkan
windows
wsl
No Label pull-request
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
GiteaMirror ninjasurge
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#60405
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?