github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 00:22:43 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[PR #9896] [MERGED] Pass input tensor instead of raw data to models #18358

New Issue
Closed
opened 2026-04-16 06:32:47 -05:00 by GiteaMirror · 0 comments
GiteaMirror commented 2026-04-16 06:32:47 -05:00
Owner
Copy Link

📋 Pull Request Information

Original PR: https://github.com/ollama/ollama/pull/9896
Author: @jessegross
Created: 3/19/2025
Status: ✅ Merged
Merged: 3/20/2025
Merged by: @jessegross

Base: main ← Head: jessegross/inputs

📝 Commits (2)

  • 1d87a4a input: Rename Options to Batch
  • a310fa7 model: Pass input tensor instead of raw data to models

📊 Changes

14 files changed (+81 additions, -79 deletions)

View changed files

📝 kvcache/cache.go (+1 -1)
📝 kvcache/causal.go (+6 -6)
📝 kvcache/causal_test.go (+1 -1)
📝 kvcache/encoder.go (+3 -3)
📝 kvcache/wrapper.go (+4 -4)
📝 model/input/input.go (+21 -6)
📝 model/model.go (+13 -7)
📝 model/model_test.go (+1 -1)
📝 model/models/gemma2/model.go (+4 -9)
📝 model/models/gemma3/model.go (+4 -9)
📝 model/models/gemma3/model_text.go (+2 -2)
📝 model/models/llama/model.go (+4 -9)
📝 model/models/mllama/model.go (+6 -11)
📝 runner/ollamarunner/runner.go (+11 -10)

📄 Description

Rather than directly giving the input data to models, we can pass a tensor instead. In the short term, this saves some duplicated code.

Longer term, we will want to overlap setting up the next batch with processing of the current one. In this case, we will only have the shape of the tensor but it will not be loaded with data at the time of graph generation. By passing only a tensor to models now, we set up this possibility and prevent them from relying on data that they won't have in the future.

Although the same could be done for Positions and Outputs, in some cases we either need the raw input data or don't use them at all. Therefore, for now we leave them as they are and allow models to convert them to tensors as needed.

🔄 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/9896 **Author:** [@jessegross](https://github.com/jessegross) **Created:** 3/19/2025 **Status:** ✅ Merged **Merged:** 3/20/2025 **Merged by:** [@jessegross](https://github.com/jessegross) **Base:** `main` ← **Head:** `jessegross/inputs` --- ### 📝 Commits (2) - [`1d87a4a`](https://github.com/ollama/ollama/commit/1d87a4a2e095deb9900ed9a69a46cd355125eb66) input: Rename Options to Batch - [`a310fa7`](https://github.com/ollama/ollama/commit/a310fa7ec67ebd68cd8cafaf96a16fec110569f2) model: Pass input tensor instead of raw data to models ### 📊 Changes **14 files changed** (+81 additions, -79 deletions) <details> <summary>View changed files</summary> 📝 `kvcache/cache.go` (+1 -1) 📝 `kvcache/causal.go` (+6 -6) 📝 `kvcache/causal_test.go` (+1 -1) 📝 `kvcache/encoder.go` (+3 -3) 📝 `kvcache/wrapper.go` (+4 -4) 📝 `model/input/input.go` (+21 -6) 📝 `model/model.go` (+13 -7) 📝 `model/model_test.go` (+1 -1) 📝 `model/models/gemma2/model.go` (+4 -9) 📝 `model/models/gemma3/model.go` (+4 -9) 📝 `model/models/gemma3/model_text.go` (+2 -2) 📝 `model/models/llama/model.go` (+4 -9) 📝 `model/models/mllama/model.go` (+6 -11) 📝 `runner/ollamarunner/runner.go` (+11 -10) </details> ### 📄 Description Rather than directly giving the input data to models, we can pass a tensor instead. In the short term, this saves some duplicated code. Longer term, we will want to overlap setting up the next batch with processing of the current one. In this case, we will only have the shape of the tensor but it will not be loaded with data at the time of graph generation. By passing only a tensor to models now, we set up this possibility and prevent them from relying on data that they won't have in the future. Although the same could be done for Positions and Outputs, in some cases we either need the raw input data or don't use them at all. Therefore, for now we leave them as they are and allow models to convert them to tensors as needed. --- <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 2026-04-16 06:32:47 -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#18358
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?