github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-06 08:02:14 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #11193] {"error":"\"llava:7b\" does not support chat"} #7375

New Issue
Closed
opened 2026-04-12 19:25:59 -05:00 by GiteaMirror · 8 comments
GiteaMirror commented 2026-04-12 19:25:59 -05:00
Owner
Copy Link

Originally created by @fishfl on GitHub (Jun 25, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/11193

What is the issue?

It's not like #11114 , I have re-test on the latest version 0.9.2 of ollama, it still error.

We try to load the multi-modal Model:llava:7b use the API:
curl http://localhost:11434/api/generate -d '{"model":"llava:7b"}'

The ollama response as follows:
{"error":""llava:7b" does not support generate"}

The same error use the CLI 'ollama run' command.

logs:
Couldn't find '/root/.ollama/id_ed25519'. Generating new private key.
Your new public key is:

ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIKPH5BD/eYQhgy15DLVLKAheo4hGi9SySAMKuqO86OTI

time=2025-06-25T02:57:48.617Z level=INFO source=routes.go:1235 msg="server config" env="map[CUDA_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_V
ERSION: HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_CONTEXT_LENGTH:4096 OLLAMA_DEBUG:INFO OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://0.0.0.0:11434 O
LLAMA_INTEL_GPU:false OLLAMA_KEEP_ALIVE:5m0s OLLAMA_KV_CACHE_TYPE: OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:0 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS
:/root/.ollama/models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NUM_PARALLEL:0 OLLAMA_ORIGINS:[http://localhost
https://localhost http://localhost:* https://localhost:* http://127.0.0.1 https://127.0.0.1 http://127.0.0.1:* https://127.0.0.1:* http://0.0.0.0 https://0.0.0.0 http://0.0
.0.0:* https://0.0.0.0:* app://* file://* tauri://* vscode-webview://* vscode-file://*] OLLAMA_SCHED_SPREAD:false ROCR_VISIBLE_DEVICES: http_proxy: https_proxy: no_proxy:]"
time=2025-06-25T02:57:48.617Z level=INFO source=images.go:480 msg="total blobs: 0"
time=2025-06-25T02:57:48.617Z level=INFO source=images.go:487 msg="total unused blobs removed: 0"
time=2025-06-25T02:57:48.617Z level=INFO source=routes.go:1288 msg="Listening on [::]:11434 (version 0.9.2)"
time=2025-06-25T02:57:48.617Z level=INFO source=gpu.go:217 msg="looking for compatible GPUs"
time=2025-06-25T02:57:48.640Z level=INFO source=gpu.go:377 msg="no compatible GPUs were discovered"
time=2025-06-25T02:57:48.640Z level=INFO source=types.go:130 msg="inference compute" id=0 library=cpu variant="" compute="" driver=0.0 name="" total="503.0 GiB" available="2
22.3 GiB"
time=2025-06-25T03:00:34.774Z level=ERROR source=images.go:91 msg="couldn't decode ggml" error=EOF
[GIN] 2025/06/25 - 03:00:34 | 400 | 986.303µs | 10.133.121.193 | POST "/api/chat"
[GIN] 2025/06/25 - 03:01:35 | 200 | 46.789µs | 127.0.0.1 | GET "/api/version"
[GIN] 2025/06/25 - 03:01:39 | 200 | 31.18µs | 127.0.0.1 | HEAD "/"
[GIN] 2025/06/25 - 03:01:39 | 200 | 341.067µs | 127.0.0.1 | GET "/api/tags"

A little views:
1、 only multi-modal model not work (llava / qwen2.5vl). The text ONLY LLM work well.
2、 This might be related to the Libs of linux.

uname -a
Linux furt--bjhotel-search-rankv5-pri-ms-tf--29276 5.15.146-1.el7.x86_64 #1 SMP Wed Jan 10 20:31:02 CST 2024 x86_64 x86_64 x86_64 GNU/Linux

Thanks.

Relevant log output

OS

Linux

GPU

No response

CPU

Intel

Ollama version

0.9.2

Originally created by @fishfl on GitHub (Jun 25, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/11193 ### What is the issue? It's not like #11114 , I have re-test on the latest version 0.9.2 of ollama, it still error. We try to load the multi-modal Model:llava:7b use the API: curl http://localhost:11434/api/generate -d '{"model":"llava:7b"}' The ollama response as follows: {"error":""llava:7b" does not support generate"} The same error use the CLI 'ollama run' command. logs: Couldn't find '/root/.ollama/id_ed25519'. Generating new private key. Your new public key is: ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIKPH5BD/eYQhgy15DLVLKAheo4hGi9SySAMKuqO86OTI time=2025-06-25T02:57:48.617Z level=INFO source=routes.go:1235 msg="server config" env="map[CUDA_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_V ERSION: HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_CONTEXT_LENGTH:4096 OLLAMA_DEBUG:INFO OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://0.0.0.0:11434 O LLAMA_INTEL_GPU:false OLLAMA_KEEP_ALIVE:5m0s OLLAMA_KV_CACHE_TYPE: OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:0 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS :/root/.ollama/models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NUM_PARALLEL:0 OLLAMA_ORIGINS:[http://localhost https://localhost http://localhost:* https://localhost:* http://127.0.0.1 https://127.0.0.1 http://127.0.0.1:* https://127.0.0.1:* http://0.0.0.0 https://0.0.0.0 http://0.0 .0.0:* https://0.0.0.0:* app://* file://* tauri://* vscode-webview://* vscode-file://*] OLLAMA_SCHED_SPREAD:false ROCR_VISIBLE_DEVICES: http_proxy: https_proxy: no_proxy:]" time=2025-06-25T02:57:48.617Z level=INFO source=images.go:480 msg="total blobs: 0" time=2025-06-25T02:57:48.617Z level=INFO source=images.go:487 msg="total unused blobs removed: 0" time=2025-06-25T02:57:48.617Z level=INFO source=routes.go:1288 msg="Listening on [::]:11434 (version 0.9.2)" time=2025-06-25T02:57:48.617Z level=INFO source=gpu.go:217 msg="looking for compatible GPUs" time=2025-06-25T02:57:48.640Z level=INFO source=gpu.go:377 msg="no compatible GPUs were discovered" time=2025-06-25T02:57:48.640Z level=INFO source=types.go:130 msg="inference compute" id=0 library=cpu variant="" compute="" driver=0.0 name="" total="503.0 GiB" available="2 22.3 GiB" time=2025-06-25T03:00:34.774Z level=ERROR source=images.go:91 msg="couldn't decode ggml" error=EOF [GIN] 2025/06/25 - 03:00:34 | 400 | 986.303µs | 10.133.121.193 | POST "/api/chat" [GIN] 2025/06/25 - 03:01:35 | 200 | 46.789µs | 127.0.0.1 | GET "/api/version" [GIN] 2025/06/25 - 03:01:39 | 200 | 31.18µs | 127.0.0.1 | HEAD "/" [GIN] 2025/06/25 - 03:01:39 | 200 | 341.067µs | 127.0.0.1 | GET "/api/tags" A little views: 1、 only multi-modal model not work (llava / qwen2.5vl). The text ONLY LLM work well. 2、 This might be related to the Libs of linux. uname -a Linux furt--bjhotel-search-rankv5-pri-ms-tf--29276 5.15.146-1.el7.x86_64 #1 SMP Wed Jan 10 20:31:02 CST 2024 x86_64 x86_64 x86_64 GNU/Linux Thanks. ### Relevant log output ```shell ``` ### OS Linux ### GPU _No response_ ### CPU Intel ### Ollama version 0.9.2
GiteaMirror added the bug label 2026-04-12 19:25:59 -05:00
GiteaMirror commented 2026-04-12 19:26:00 -05:00
Author
Owner
Copy Link

@kris-gaudel commented on GitHub (Jun 26, 2025):

Hi I'm interested in working on this issue, @fishfl is this issue still persisting for you? If so what are the steps to reproduce?

<!-- gh-comment-id:3009073472 --> @kris-gaudel commented on GitHub (Jun 26, 2025): Hi I'm interested in working on this issue, @fishfl is this issue still persisting for you? If so what are the steps to reproduce?
GiteaMirror commented 2026-04-12 19:26:01 -05:00
Author
Owner
Copy Link

@mxyng commented on GitHub (Jun 26, 2025):

Can you confirm you model is up-to-date? If you have an older model, it might be imcompatible

$ ollama ls llava:7b
NAME        ID              SIZE      MODIFIED
llava:7b    8dd30f6b0cb1    4.7 GB    3 minutes ago
<!-- gh-comment-id:3009637215 --> @mxyng commented on GitHub (Jun 26, 2025): Can you confirm you model is up-to-date? If you have an older model, it might be imcompatible ``` $ ollama ls llava:7b NAME ID SIZE MODIFIED llava:7b 8dd30f6b0cb1 4.7 GB 3 minutes ago ```
GiteaMirror commented 2026-04-12 19:26:02 -05:00
Author
Owner
Copy Link

@fishfl commented on GitHub (Jun 27, 2025):

@mxyng yes, it's up-to-date.

Image

<!-- gh-comment-id:3012326570 --> @fishfl commented on GitHub (Jun 27, 2025): @mxyng yes, it's up-to-date. ![Image](https://github.com/user-attachments/assets/9a865f5b-08a2-4053-a8d5-07435af1b29d)
GiteaMirror commented 2026-04-12 19:26:04 -05:00
Author
Owner
Copy Link

@fishfl commented on GitHub (Jun 27, 2025):

Hi I'm interested in working on this issue, @fishfl is this issue still persisting for you? If so what are the steps to reproduce?

Yes, it's still persist.

To reproduce:

1、docker pull the 0.9.2 image.
2、ollama pull the model: llava:7b
3、zip the ~/.ollama dir and copy to another local host and unzip it.
(the local host machine is not connect to the internet. I have tried this way and the text ONLY LLM work well.)
4、ollama run llava:7b or use the REST api then you got the error.

The additional images maybe useful.

And This Might be related to the libraries of linux. Because on some host it can work. (The host machine which can pull model from internet)
On some host machine(the local machine) it's not work. But I don't know what library of linux system that ollama depends on.

On the local host machine:
uname -r
5.15.146-1.el7.x86_64

Image
Image

<!-- gh-comment-id:3012379216 --> @fishfl commented on GitHub (Jun 27, 2025): > Hi I'm interested in working on this issue, [@fishfl](https://github.com/fishfl) is this issue still persisting for you? If so what are the steps to reproduce? Yes, it's still persist. To reproduce: 1、docker pull the 0.9.2 image. 2、ollama pull the model: llava:7b 3、zip the ~/.ollama dir and copy to another local host and unzip it. (the local host machine is not connect to the internet. I have tried this way and the text ONLY LLM work well.) 4、ollama run llava:7b or use the REST api then you got the error. The additional images maybe useful. And This Might be related to the libraries of linux. Because on some host it can work. (The host machine which can pull model from internet) On some host machine(the local machine) it's not work. But I don't know what library of linux system that ollama depends on. On the local host machine: uname -r 5.15.146-1.el7.x86_64 ![Image](https://github.com/user-attachments/assets/12036a3a-3fba-4a8e-a583-12e8bc67c124) ![Image](https://github.com/user-attachments/assets/92babade-7400-4559-b1a6-7182a62a198a)
GiteaMirror commented 2026-04-12 19:26:04 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Jun 27, 2025):

What's the output of

ls -lR ~/.ollama
sha256sum ~/.ollama/models/blobs/*
<!-- gh-comment-id:3013335132 --> @rick-github commented on GitHub (Jun 27, 2025): What's the output of ``` ls -lR ~/.ollama sha256sum ~/.ollama/models/blobs/* ```
GiteaMirror commented 2026-04-12 19:26:05 -05:00
Author
Owner
Copy Link

@jmorganca commented on GitHub (Jun 27, 2025):

This should be fixed now as of 0.9.3. If you're still hitting the error we can re-open the issue and get it fixed – sorry about that!

<!-- gh-comment-id:3013517210 --> @jmorganca commented on GitHub (Jun 27, 2025): This should be fixed now as of 0.9.3. If you're still hitting the error we can re-open the issue and get it fixed – sorry about that!
GiteaMirror commented 2026-04-12 19:26:06 -05:00
Author
Owner
Copy Link

@fishfl commented on GitHub (Jun 30, 2025):

What's the output of

ls -lR ~/.ollama
sha256sum ~/.ollama/models/blobs/*

root@furt--bjhotel-search-rankv5-pri-ms-tf--29276:/home/work# ls -lR ~/.ollama
/root/.ollama:
total 8
-rw------- 1 root root 387 Jun 25 02:57 id_ed25519
-rw-r--r-- 1 root root 81 Jun 25 02:57 id_ed25519.pub
drwxr-xr-x 4 root root 48 Jun 25 03:00 models

/root/.ollama/models:
total 4
drwxr-xr-x 2 root root 4096 Jun 25 03:00 blobs
drwxr-xr-x 3 root root 40 Jun 25 03:00 manifests

/root/.ollama/models/blobs:
total 609824
-rw-r--r-- 1 root root 0 Jun 25 03:00 sha256-170370233dd5c5415250a2ecd5c71586352850729062ccef1496385647293868
-rw-r--r-- 1 root root 11356 Jun 25 03:00 sha256-43070e2d4e532684de521b885f385d0841030efa2b1a20bafb76133a5e1379c1
-rw-r--r-- 1 root root 624434368 Jun 25 03:00 sha256-72d6f08a42f656d36b356dbe0920675899a99ce21192fd66266fb7d82ed07539
-rw-r--r-- 1 root root 564 Jun 25 03:00 sha256-7c658f9561e5dbbafb042a00f6a4de57877adddd957809111f3123e272632b4d
-rw-r--r-- 1 root root 67 Jun 25 03:00 sha256-c43332387573e98fdfad4a606171279955b53d891ba2500552c2984a6560ffb4
-rw-r--r-- 1 root root 30 Jun 25 03:00 sha256-ed11eda7790d05b49395598a42b155812b17e263214292f7b87d15e14003d337

/root/.ollama/models/manifests:
total 0
drwxr-xr-x 3 root root 21 Jun 25 03:00 registry.ollama.ai

/root/.ollama/models/manifests/registry.ollama.ai:
total 0
drwxr-xr-x 3 root root 19 Jun 25 03:00 library

/root/.ollama/models/manifests/registry.ollama.ai/library:
total 0
drwxr-xr-x 2 root root 16 Jun 25 03:00 llava

/root/.ollama/models/manifests/registry.ollama.ai/library/llava:
total 4
-rw-r--r-- 1 root root 1011 Jun 25 03:00 7b

root@furt--bjhotel-search-rankv5-pri-ms-tf--29276:/home/work# sha256sum ~/.ollama/models/blobs/*
e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855 /root/.ollama/models/blobs/sha256-170370233dd5c5415250a2ecd5c71586352850729062ccef1496385647293868
43070e2d4e532684de521b885f385d0841030efa2b1a20bafb76133a5e1379c1 /root/.ollama/models/blobs/sha256-43070e2d4e532684de521b885f385d0841030efa2b1a20bafb76133a5e1379c1
72d6f08a42f656d36b356dbe0920675899a99ce21192fd66266fb7d82ed07539 /root/.ollama/models/blobs/sha256-72d6f08a42f656d36b356dbe0920675899a99ce21192fd66266fb7d82ed07539
7c658f9561e5dbbafb042a00f6a4de57877adddd957809111f3123e272632b4d /root/.ollama/models/blobs/sha256-7c658f9561e5dbbafb042a00f6a4de57877adddd957809111f3123e272632b4d
c43332387573e98fdfad4a606171279955b53d891ba2500552c2984a6560ffb4 /root/.ollama/models/blobs/sha256-c43332387573e98fdfad4a606171279955b53d891ba2500552c2984a6560ffb4
ed11eda7790d05b49395598a42b155812b17e263214292f7b87d15e14003d337 /root/.ollama/models/blobs/sha256-ed11eda7790d05b49395598a42b155812b17e263214292f7b87d15e14003d337
root@furt--bjhotel-search-rankv5-pri-ms-tf--29276:/home/work#

<!-- gh-comment-id:3018451109 --> @fishfl commented on GitHub (Jun 30, 2025): > What's the output of > > ``` > ls -lR ~/.ollama > sha256sum ~/.ollama/models/blobs/* > ``` root@furt--bjhotel-search-rankv5-pri-ms-tf--29276:/home/work# ls -lR ~/.ollama /root/.ollama: total 8 -rw------- 1 root root 387 Jun 25 02:57 id_ed25519 -rw-r--r-- 1 root root 81 Jun 25 02:57 id_ed25519.pub drwxr-xr-x 4 root root 48 Jun 25 03:00 models /root/.ollama/models: total 4 drwxr-xr-x 2 root root 4096 Jun 25 03:00 blobs drwxr-xr-x 3 root root 40 Jun 25 03:00 manifests /root/.ollama/models/blobs: total 609824 -rw-r--r-- 1 root root 0 Jun 25 03:00 sha256-170370233dd5c5415250a2ecd5c71586352850729062ccef1496385647293868 -rw-r--r-- 1 root root 11356 Jun 25 03:00 sha256-43070e2d4e532684de521b885f385d0841030efa2b1a20bafb76133a5e1379c1 -rw-r--r-- 1 root root 624434368 Jun 25 03:00 sha256-72d6f08a42f656d36b356dbe0920675899a99ce21192fd66266fb7d82ed07539 -rw-r--r-- 1 root root 564 Jun 25 03:00 sha256-7c658f9561e5dbbafb042a00f6a4de57877adddd957809111f3123e272632b4d -rw-r--r-- 1 root root 67 Jun 25 03:00 sha256-c43332387573e98fdfad4a606171279955b53d891ba2500552c2984a6560ffb4 -rw-r--r-- 1 root root 30 Jun 25 03:00 sha256-ed11eda7790d05b49395598a42b155812b17e263214292f7b87d15e14003d337 /root/.ollama/models/manifests: total 0 drwxr-xr-x 3 root root 21 Jun 25 03:00 registry.ollama.ai /root/.ollama/models/manifests/registry.ollama.ai: total 0 drwxr-xr-x 3 root root 19 Jun 25 03:00 library /root/.ollama/models/manifests/registry.ollama.ai/library: total 0 drwxr-xr-x 2 root root 16 Jun 25 03:00 llava /root/.ollama/models/manifests/registry.ollama.ai/library/llava: total 4 -rw-r--r-- 1 root root 1011 Jun 25 03:00 7b root@furt--bjhotel-search-rankv5-pri-ms-tf--29276:/home/work# sha256sum ~/.ollama/models/blobs/* e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855 /root/.ollama/models/blobs/sha256-170370233dd5c5415250a2ecd5c71586352850729062ccef1496385647293868 43070e2d4e532684de521b885f385d0841030efa2b1a20bafb76133a5e1379c1 /root/.ollama/models/blobs/sha256-43070e2d4e532684de521b885f385d0841030efa2b1a20bafb76133a5e1379c1 72d6f08a42f656d36b356dbe0920675899a99ce21192fd66266fb7d82ed07539 /root/.ollama/models/blobs/sha256-72d6f08a42f656d36b356dbe0920675899a99ce21192fd66266fb7d82ed07539 7c658f9561e5dbbafb042a00f6a4de57877adddd957809111f3123e272632b4d /root/.ollama/models/blobs/sha256-7c658f9561e5dbbafb042a00f6a4de57877adddd957809111f3123e272632b4d c43332387573e98fdfad4a606171279955b53d891ba2500552c2984a6560ffb4 /root/.ollama/models/blobs/sha256-c43332387573e98fdfad4a606171279955b53d891ba2500552c2984a6560ffb4 ed11eda7790d05b49395598a42b155812b17e263214292f7b87d15e14003d337 /root/.ollama/models/blobs/sha256-ed11eda7790d05b49395598a42b155812b17e263214292f7b87d15e14003d337 root@furt--bjhotel-search-rankv5-pri-ms-tf--29276:/home/work#
GiteaMirror commented 2026-04-12 19:26:06 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Jun 30, 2025):

/root/.ollama/models/blobs:
total 609824
-rw-r--r-- 1 root root 0 Jun 25 03:00 sha256-170370233dd5c5415250a2ecd5c71586352850729062ccef1496385647293868

This file is broken. Re-pull the model.

<!-- gh-comment-id:3018700900 --> @rick-github commented on GitHub (Jun 30, 2025): ``` /root/.ollama/models/blobs: total 609824 -rw-r--r-- 1 root root 0 Jun 25 03:00 sha256-170370233dd5c5415250a2ecd5c71586352850729062ccef1496385647293868 ``` This file is broken. Re-pull the model.
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#7375
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?