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 #11283] Ollama 0.9.3 truncating prompt to 8192 no mater params passed #33201

New Issue
Closed
opened 2026-04-22 15:38:24 -05:00 by GiteaMirror · 2 comments
GiteaMirror commented 2026-04-22 15:38:24 -05:00
Owner
Copy Link

Originally created by @alexveli1 on GitHub (Jul 3, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/11283

What is the issue?

Dear colleagues,
Want to share experience with 0.9.3 and appreciate fixing.
I've upgraded ollama to 0.9.3 to pull and run gemma3n (unsuccessful though - gemma3n not running - but this is not a topic here - see codebase is being adapted for this model, so will wait).
The issue: Running other models on 0.9.3 (e.g. gemma3:12b-it-q8_0) with whatever OLLAMA_CONTEXT_LENGTH (100000, 70000, 10000, etc.) on ollama 0.9.3 shows that ollama truncates prompt to 8192.
My solution: Falling back to 0.9.2 solved issue with truncating prompt/context, but of course gemma3n not starting.

Environment:
OS: Ubuntu 24.04
hardware: 1xRTX 4090
docker: 28.2.2
docker compose: v2.36.2
client: open-webui 0.6.15

docker-compose.yml

 ollama:
    image: ollama/ollama:0.9.3
    volumes:
      - /home/ollama/.ollama:/root/.ollama
    container_name: ollama
    hostname: ollama
    pull_policy: refresh
    tty: true
    restart: unless-stopped
    environment:
      - OLLAMA_HOST=ollama:11434
      - OLLAMA_ORIGINS=*
      - OLLAMA_MAX_LOADED_MODELS=2
      - OLLAMA_KEEP_ALIVE=-1
      - OLLAMA_NUM_PARALLEL=1
      - OLLAMA_NUM_THREADS=10
      - OLLAMA_MAX_QUEUE=1
      - OLLAMA_NO_PULL=1
      - OLLAMA_CUDA=1
      - OLLAMA_MODELS=/root/.ollama/models
      - OLLAMA_HOST=ollama
      - OLLAMA_DEBUG=2
    ports:
      - "11434:11434"
    logging:
      driver: "json-file"
      options:
        max-size: "10m"
        max-file: "3"
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: all
              capabilities:
                - gpu

Relevant log output

msg="truncating input prompt" limit=8192 prompt=66476 keep=0 new=8192
66476 - prompt/context size.

OS

Ubuntu 24.04

GPU

RTX 4090

CPU

i9-14900K

Ollama version

0.9.3

Originally created by @alexveli1 on GitHub (Jul 3, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/11283 ### What is the issue? Dear colleagues, Want to share experience with 0.9.3 and appreciate fixing. I've upgraded ollama to 0.9.3 to pull and run gemma3n (unsuccessful though - gemma3n not running - but this is not a topic here - see codebase is being adapted for this model, so will wait). **The issue:** Running other models on 0.9.3 (e.g. gemma3:12b-it-q8_0) with whatever OLLAMA_CONTEXT_LENGTH (100000, 70000, 10000, etc.) on ollama 0.9.3 shows that ollama truncates prompt to 8192. **My solution:** Falling back to 0.9.2 solved issue with truncating prompt/context, but of course gemma3n not starting. **Environment:** **OS**: Ubuntu 24.04 **hardware**: 1xRTX 4090 **docker**: 28.2.2 **docker compose**: v2.36.2 **client**: open-webui 0.6.15 **docker-compose.yml** ```docker ollama: image: ollama/ollama:0.9.3 volumes: - /home/ollama/.ollama:/root/.ollama container_name: ollama hostname: ollama pull_policy: refresh tty: true restart: unless-stopped environment: - OLLAMA_HOST=ollama:11434 - OLLAMA_ORIGINS=* - OLLAMA_MAX_LOADED_MODELS=2 - OLLAMA_KEEP_ALIVE=-1 - OLLAMA_NUM_PARALLEL=1 - OLLAMA_NUM_THREADS=10 - OLLAMA_MAX_QUEUE=1 - OLLAMA_NO_PULL=1 - OLLAMA_CUDA=1 - OLLAMA_MODELS=/root/.ollama/models - OLLAMA_HOST=ollama - OLLAMA_DEBUG=2 ports: - "11434:11434" logging: driver: "json-file" options: max-size: "10m" max-file: "3" deploy: resources: reservations: devices: - driver: nvidia count: all capabilities: - gpu ``` ### Relevant log output ```shell msg="truncating input prompt" limit=8192 prompt=66476 keep=0 new=8192 66476 - prompt/context size. ``` ### OS Ubuntu 24.04 ### GPU RTX 4090 ### CPU i9-14900K ### Ollama version 0.9.3
GiteaMirror added the bug label 2026-04-22 15:38:24 -05:00
GiteaMirror commented 2026-04-22 15:38:26 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Jul 3, 2025):

https://github.com/ollama/ollama/pull/11175

The initial version of gemma3 had the wrong context size, so the PR above is limiting the prompt to the incorrect smaller size. Re-pull the model and the context limit will increase.

<!-- gh-comment-id:3031488259 --> @rick-github commented on GitHub (Jul 3, 2025): https://github.com/ollama/ollama/pull/11175 The initial version of gemma3 had the [wrong](https://github.com/ollama/ollama/issues/9702#issuecomment-2727063304) context size, so the PR above is limiting the prompt to the incorrect smaller size. Re-pull the model and the context limit will increase.
GiteaMirror commented 2026-04-22 15:38:27 -05:00
Author
Owner
Copy Link

@pdevine commented on GitHub (Jul 3, 2025):

@alexveli1 sorry about this. As @rick-github (thank you Rick!) mentioned, just re-pull the model and it should be fine.

<!-- gh-comment-id:3032849510 --> @pdevine commented on GitHub (Jul 3, 2025): @alexveli1 sorry about this. As @rick-github (thank you Rick!) mentioned, just re-pull the model and it should be fine.
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#33201
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?