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 #6927] Why Is n_ctx in log Always Four Times the num_ctx Value in ModelFIle When Building qwen2.5-coder-7b-instruct-q5_k_m.gguf? #30144

New Issue
Closed
opened 2026-04-22 09:37:38 -05:00 by GiteaMirror · 1 comment
GiteaMirror commented 2026-04-22 09:37:38 -05:00
Owner
Copy Link

Originally created by @XiongDaowen on GitHub (Sep 24, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/6927

What is the issue?

When I built qwen2.5-coder-7b-instruct-q5_k_m.gguf using the modelfile and set PARAMETER num_ctx 4096, the log output showed llama_new_context_with_model: n_ctx = 16384. After setting num_ctx to different values, I noticed that n_ctx is always 4 times the value of num_ctx. Why is this happening?
The log:
89d1ee67-30d0-49b6-9553-32fdc1840fb7
The ModelFile:
4dc02651-9a6d-4007-95be-adcbd9aa871e

OS

No response

GPU

No response

CPU

No response

Ollama version

0.3.3

Originally created by @XiongDaowen on GitHub (Sep 24, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/6927 ### What is the issue? When I built qwen2.5-coder-7b-instruct-q5_k_m.gguf using the modelfile and set PARAMETER num_ctx 4096, the log output showed llama_new_context_with_model: n_ctx = 16384. After setting num_ctx to different values, I noticed that n_ctx is always 4 times the value of num_ctx. Why is this happening? The log: ![89d1ee67-30d0-49b6-9553-32fdc1840fb7](https://github.com/user-attachments/assets/b9b5a0d0-727b-4422-a592-320784054634) The ModelFile: ![4dc02651-9a6d-4007-95be-adcbd9aa871e](https://github.com/user-attachments/assets/591e6e7c-3047-478e-9207-19c2fb3ea509) ### OS _No response_ ### GPU _No response_ ### CPU _No response_ ### Ollama version 0.3.3
GiteaMirror added the bug label 2026-04-22 09:37:38 -05:00
GiteaMirror commented 2026-04-22 09:37:39 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Sep 24, 2024):

Either OLLAMA_NUM_PARALLEL is unset or set to 4. OLLAMA_NUM_PARALLEL sets how many requests ollama can handle concurrently. Each request handler needs its own context space, so total context is (number of handlers * context size). If OLLAMA_NUM_PARALLEL is unset, ollama chooses either 1 or 4 depending on how much VRAM is available. If you set OLLAMA_NUM_PARALLEL=1 in the server environment, n_ctx will be 4096.

<!-- gh-comment-id:2370278545 --> @rick-github commented on GitHub (Sep 24, 2024): Either `OLLAMA_NUM_PARALLEL` is unset or set to 4. `OLLAMA_NUM_PARALLEL` sets how many requests ollama can handle concurrently. Each request handler needs its own context space, so total context is (number of handlers * context size). If `OLLAMA_NUM_PARALLEL` is unset, ollama chooses either 1 or 4 depending on how much VRAM is available. If you set `OLLAMA_NUM_PARALLEL=1` in the [server environment](https://github.com/ollama/ollama/blob/main/docs/faq.md#how-do-i-configure-ollama-server), `n_ctx` will be 4096.
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 bug
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#30144
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?