github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 08:30:05 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #15828] qwen2.5vl:7b GGML_ASSERT(a->ne[2] * 4 == b->ne[0]) - 500 Internal Server Error on vision inference #56600

New Issue
Open
opened 2026-04-29 11:05:03 -05:00 by GiteaMirror · 0 comments
GiteaMirror commented 2026-04-29 11:05:03 -05:00
Owner
Copy Link

Originally created by @tustuntas on GitHub (Apr 26, 2026).
Original GitHub issue: https://github.com/ollama/ollama/issues/15828

What is the issue?

Body:

Environment

  • OS: Windows 11
  • GPU: NVIDIA RTX 4090 24GB VRAM, Driver 591.86, CUDA 13.1
  • Ollama: 0.21.3-rc0
  • Model: qwen2.5vl:7b (ID: 5ced39dfa4ba, 6.0 GB, 29/29 layers on GPU)

Description

During batch vision inference (PDF page-by-page extraction with Qwen2.5-VL), Ollama randomly returns 500 Internal Server Error. The server log shows:

GGML_ASSERT(a->ne[2] * 4 == b->ne[0])

The error is not deterministic — the same image succeeds on retry or after an Ollama restart. During sustained batch processing (~10-15 sequential requests), the error rate is approximately 30-40%.

Environment Variables

OLLAMA_FLASH_ATTENTION=1
OLLAMA_NUM_GPU=99
OLLAMA_NUM_PARALLEL=1

Steps to Reproduce

  1. Load qwen2.5vl:7b model
  2. Send 10-15 sequential vision requests with 150 DPI page images (1288x1638 pixels)
  3. Observe intermittent 500 errors on random pages

Error Pattern

  • First few pages typically succeed
  • After 3-5 successful requests, 500 errors start appearing
  • Error rate increases with sustained usage
  • Restarting Ollama temporarily clears the issue
  • Splitting images into smaller halves (1288x839) reduces error frequency

Workaround

  • Splitting the image into 2 halves before sending reduces error rate
  • Restarting Ollama service after consecutive failures
  • Using keep_alive and retry logic in the client

Model Details

NAME                       ID              SIZE
qwen2.5vl:7b               5ced39dfa4ba    6.0 GB
qwen2.5vl:32b              3edc3a52fe98    21 GB

Same issue also occurs with qwen2.5vl:32b (49/65 layers on GPU) but less frequently due to slower processing.


### Relevant log output

```shell

OS

Windows

GPU

Nvidia

CPU

Intel

Ollama version

0.21.3-rc0

Originally created by @tustuntas on GitHub (Apr 26, 2026). Original GitHub issue: https://github.com/ollama/ollama/issues/15828 ### What is the issue? **Body:** ### Environment - **OS:** Windows 11 - **GPU:** NVIDIA RTX 4090 24GB VRAM, Driver 591.86, CUDA 13.1 - **Ollama:** 0.21.3-rc0 - **Model:** qwen2.5vl:7b (ID: 5ced39dfa4ba, 6.0 GB, 29/29 layers on GPU) ### Description During batch vision inference (PDF page-by-page extraction with Qwen2.5-VL), Ollama randomly returns `500 Internal Server Error`. The server log shows: ``` GGML_ASSERT(a->ne[2] * 4 == b->ne[0]) ``` The error is **not deterministic** — the same image succeeds on retry or after an Ollama restart. During sustained batch processing (~10-15 sequential requests), the error rate is approximately 30-40%. ### Environment Variables ``` OLLAMA_FLASH_ATTENTION=1 OLLAMA_NUM_GPU=99 OLLAMA_NUM_PARALLEL=1 ``` ### Steps to Reproduce 1. Load `qwen2.5vl:7b` model 2. Send 10-15 sequential vision requests with 150 DPI page images (1288x1638 pixels) 3. Observe intermittent 500 errors on random pages ### Error Pattern - First few pages typically succeed - After 3-5 successful requests, 500 errors start appearing - Error rate increases with sustained usage - Restarting Ollama temporarily clears the issue - Splitting images into smaller halves (1288x839) reduces error frequency ### Workaround - Splitting the image into 2 halves before sending reduces error rate - Restarting Ollama service after consecutive failures - Using `keep_alive` and retry logic in the client ### Model Details ``` NAME ID SIZE qwen2.5vl:7b 5ced39dfa4ba 6.0 GB qwen2.5vl:32b 3edc3a52fe98 21 GB ``` Same issue also occurs with `qwen2.5vl:32b` (49/65 layers on GPU) but less frequently due to slower processing. ``` ### Relevant log output ```shell ``` ### OS Windows ### GPU Nvidia ### CPU Intel ### Ollama version 0.21.3-rc0
GiteaMirror added the bug label 2026-04-29 11:05:03 -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 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#56600
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?