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

[GH-ISSUE #11277] Parallel Computing Support for Concurrent Ollama Requests #69495

New Issue
Closed
opened 2026-05-04 18:16:01 -05:00 by GiteaMirror · 1 comment
GiteaMirror commented 2026-05-04 18:16:01 -05:00
Owner
Copy Link

Originally created by @pnwttklm on GitHub (Jul 2, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/11277

When running two separate Python programs that use the same local Ollama endpoint (e.g., http://localhost:11434), both programs experience significant slowdowns. It appears that Ollama does not efficiently handle concurrent requests — possibly due to a lack of parallel execution or limited context switching.

This makes it challenging to use Ollama in multi-process or multi-threaded environments such as:
• Chat applications with multiple sessions
• Background workers
• Cron jobs running alongside servers
• Simultaneous local clients (e.g., API + CLI)

Steps to reproduce
1. Start the Ollama server: ollama run llama3
2. Open two terminals
3. Run two Python scripts that both make POST requests to /api/generate
4. Observe degraded performance in both — response times increase significantly

Environtment
OS: Windows 11 with WSL2 (Ubuntu)
CPU: Intel Core i7-12700K (12th Gen, 20 threads)
GPU: NVIDIA GeForce RTX 4090
Ollama Version: 0.9.5
Model Used: llama3, gemma3,
Python Version: 3.12 (via WSL2)

Originally created by @pnwttklm on GitHub (Jul 2, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/11277 When running two separate Python programs that use the same local Ollama endpoint (e.g., http://localhost:11434), both programs experience significant slowdowns. It appears that Ollama does not efficiently handle concurrent requests — possibly due to a lack of parallel execution or limited context switching. This makes it challenging to use Ollama in multi-process or multi-threaded environments such as: • Chat applications with multiple sessions • Background workers • Cron jobs running alongside servers • Simultaneous local clients (e.g., API + CLI) Steps to reproduce 1. Start the Ollama server: ollama run llama3 2. Open two terminals 3. Run two Python scripts that both make POST requests to /api/generate 4. Observe degraded performance in both — response times increase significantly Environtment OS: Windows 11 with WSL2 (Ubuntu) CPU: Intel Core i7-12700K (12th Gen, 20 threads) GPU: NVIDIA GeForce RTX 4090 Ollama Version: 0.9.5 Model Used: llama3, gemma3, Python Version: 3.12 (via WSL2)
GiteaMirror added the feature requestneeds more info labels 2026-05-04 18:16:01 -05:00
GiteaMirror commented 2026-05-04 18:16:03 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Jul 2, 2025):

Have you set OLLAMA_NUM_PARALLEL?

<!-- gh-comment-id:3029448184 --> @rick-github commented on GitHub (Jul 2, 2025): Have you set [`OLLAMA_NUM_PARALLEL`](https://github.com/ollama/ollama/blob/main/docs/faq.md#how-does-ollama-handle-concurrent-requests)?
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 feature request needs more info
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#69495
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?