github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-06 16:11:34 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #845] Ollama Docker: Error LLama runner process has terminated #62440

New Issue
Closed
opened 2026-05-03 08:58:08 -05:00 by GiteaMirror · 42 comments
GiteaMirror commented 2026-05-03 08:58:08 -05:00
Owner
Copy Link

Originally created by @randywreed on GitHub (Oct 19, 2023).
Original GitHub issue: https://github.com/ollama/ollama/issues/845

I'm running the latest docker version of ollama (as of 10/19/2023). When I do docker exec -it ollama ollama run mistral
I get the error Error Llama runner process has terminated.
The docker does not have .ollama/logs directory and journalctl is not installed.
Inside the docker it seems to have plenty of space, and free -m reports it has 127gb of ram available

Any help would be appreciated.

Originally created by @randywreed on GitHub (Oct 19, 2023). Original GitHub issue: https://github.com/ollama/ollama/issues/845 I'm running the latest docker version of ollama (as of 10/19/2023). When I do docker exec -it ollama ollama run mistral I get the error Error Llama runner process has terminated. The docker does not have .ollama/logs directory and journalctl is not installed. Inside the docker it seems to have plenty of space, and free -m reports it has 127gb of ram available Any help would be appreciated.
GiteaMirror commented 2026-05-03 08:58:09 -05:00
Author
Owner
Copy Link

@mxyng commented on GitHub (Oct 19, 2023):

Logs for Docker Ollama can be retrieved with docker logs ollama. Please attach it for troubleshooting. It's possible it's related to unsupported CPU instructions similar to https://github.com/jmorganca/ollama/pull/778 but it's impossible to know without logs.

What platform is this using?

<!-- gh-comment-id:1771776587 --> @mxyng commented on GitHub (Oct 19, 2023): Logs for Docker Ollama can be retrieved with `docker logs ollama`. Please attach it for troubleshooting. It's possible it's related to unsupported CPU instructions similar to https://github.com/jmorganca/ollama/pull/778 but it's impossible to know without logs. What platform is this using?
GiteaMirror commented 2026-05-03 08:58:10 -05:00
Author
Owner
Copy Link

@randywreed commented on GitHub (Oct 19, 2023):

Thanks here it is. The docker is running on a linux machine, running centos. #778 was merged 3 days ago, the latest docker image is 6 days old.

2023/10/19 18:59:13 download.go:126: downloading 8daa9615cce3 in 64 59 MB part(s)
2023/10/19 18:59:48 download.go:126: downloading 8c17c2ebb0ea in 1 7 KB part(s)
2023/10/19 18:59:51 download.go:126: downloading 7c23fb36d801 in 1 4 KB part(s)
2023/10/19 18:59:54 download.go:126: downloading bec56154823a in 1 59 B part(s)
2023/10/19 18:59:57 download.go:126: downloading e35ab70a78c7 in 1 90 B part(s)
2023/10/19 18:59:59 download.go:126: downloading 09fe89200c09 in 1 529 B part(s)
[GIN] 2023/10/19 - 19:00:22 | 200 |         1m11s |       127.0.0.1 | POST     "/api/pull"
2023/10/19 19:00:23 llama.go:333: skipping accelerated runner because num_gpu=0
2023/10/19 19:00:23 llama.go:356: starting llama runner
2023/10/19 19:00:23 llama.go:408: waiting for llama runner to start responding
2023/10/19 19:00:23 llama.go:373: error starting llama runner: llama runner process has terminated
[GIN] 2023/10/19 - 19:00:23 | 500 |  506.899039ms |       127.0.0.1 | POST     "/api/generate"
2023/10/19 19:00:23 llama.go:438: llama runner stopped with error: signal: illegal instruction (core dumped)
<!-- gh-comment-id:1771831840 --> @randywreed commented on GitHub (Oct 19, 2023): Thanks here it is. The docker is running on a linux machine, running centos. #778 was merged 3 days ago, the latest docker image is 6 days old. ``` 2023/10/19 18:59:13 download.go:126: downloading 8daa9615cce3 in 64 59 MB part(s) 2023/10/19 18:59:48 download.go:126: downloading 8c17c2ebb0ea in 1 7 KB part(s) 2023/10/19 18:59:51 download.go:126: downloading 7c23fb36d801 in 1 4 KB part(s) 2023/10/19 18:59:54 download.go:126: downloading bec56154823a in 1 59 B part(s) 2023/10/19 18:59:57 download.go:126: downloading e35ab70a78c7 in 1 90 B part(s) 2023/10/19 18:59:59 download.go:126: downloading 09fe89200c09 in 1 529 B part(s) [GIN] 2023/10/19 - 19:00:22 | 200 | 1m11s | 127.0.0.1 | POST "/api/pull" 2023/10/19 19:00:23 llama.go:333: skipping accelerated runner because num_gpu=0 2023/10/19 19:00:23 llama.go:356: starting llama runner 2023/10/19 19:00:23 llama.go:408: waiting for llama runner to start responding 2023/10/19 19:00:23 llama.go:373: error starting llama runner: llama runner process has terminated [GIN] 2023/10/19 - 19:00:23 | 500 | 506.899039ms | 127.0.0.1 | POST "/api/generate" 2023/10/19 19:00:23 llama.go:438: llama runner stopped with error: signal: illegal instruction (core dumped) ```
GiteaMirror commented 2026-05-03 08:58:11 -05:00
Author
Owner
Copy Link

@mxyng commented on GitHub (Oct 19, 2023):

those are the same symptoms as #778 which we're actively investigating

<!-- gh-comment-id:1771836754 --> @mxyng commented on GitHub (Oct 19, 2023): those are the same symptoms as #778 which we're actively investigating
GiteaMirror commented 2026-05-03 08:58:12 -05:00
Author
Owner
Copy Link

@randywreed commented on GitHub (Oct 20, 2023):

Just for fun, I tried to rebuild the docker from the dockerfile in the repo, since grabs the latest update which would include the #778 pr. The error persisted

[GIN] 2023/10/20 - 19:45:33 | 200 |          1m7s |       127.0.0.1 | POST     "/api/pull"
2023/10/20 19:45:36 llama.go:340: skipping accelerated runner because num_gpu=0
2023/10/20 19:45:36 llama.go:363: starting llama runner
2023/10/20 19:45:36 llama.go:421: waiting for llama runner to start responding
2023/10/20 19:45:36 llama.go:378: signal: illegal instruction (core dumped)
2023/10/20 19:45:36 llama.go:386: error starting llama runner: llama runner process has terminated
2023/10/20 19:45:36 llama.go:452: llama runner stopped successfully
[GIN] 2023/10/20 - 19:45:36 | 500 |  2.570280569s |       127.0.0.1 | POST     "/api/generate"
<!-- gh-comment-id:1773314458 --> @randywreed commented on GitHub (Oct 20, 2023): Just for fun, I tried to rebuild the docker from the dockerfile in the repo, since grabs the latest update which would include the #778 pr. The error persisted ``` [GIN] 2023/10/20 - 19:45:33 | 200 | 1m7s | 127.0.0.1 | POST "/api/pull" 2023/10/20 19:45:36 llama.go:340: skipping accelerated runner because num_gpu=0 2023/10/20 19:45:36 llama.go:363: starting llama runner 2023/10/20 19:45:36 llama.go:421: waiting for llama runner to start responding 2023/10/20 19:45:36 llama.go:378: signal: illegal instruction (core dumped) 2023/10/20 19:45:36 llama.go:386: error starting llama runner: llama runner process has terminated 2023/10/20 19:45:36 llama.go:452: llama runner stopped successfully [GIN] 2023/10/20 - 19:45:36 | 500 | 2.570280569s | 127.0.0.1 | POST "/api/generate" ```
GiteaMirror commented 2026-05-03 08:58:14 -05:00
Author
Owner
Copy Link

@sunzh231 commented on GitHub (Oct 22, 2023):

same error in Dell R720.
CPU: Intel(R) Xeon(R) CPU E5-2670 0 @ 2.60GHz
Physical Host Operating System: vmware ESXI 6.7.0
Virtual host operating system: Ubuntu 22.04

<!-- gh-comment-id:1774033996 --> @sunzh231 commented on GitHub (Oct 22, 2023): same error in Dell R720. CPU: Intel(R) Xeon(R) CPU E5-2670 0 @ 2.60GHz Physical Host Operating System: vmware ESXI 6.7.0 Virtual host operating system: Ubuntu 22.04
GiteaMirror commented 2026-05-03 08:58:17 -05:00
Author
Owner
Copy Link

@fabceolin commented on GitHub (Oct 22, 2023):

same here:
2023/10/22 18:48:11 llama.go:356: starting llama runner
2023/10/22 18:48:11 llama.go:408: waiting for llama runner to start responding
2023/10/22 18:48:11 llama.go:373: error starting llama runner: llama runner process has terminated
2023/10/22 18:48:11 llama.go:438: llama runner stopped with error: signal: illegal instruction (core dumped)
2023/10/22 18:48:11 llama.go:356: starting llama runner
2023/10/22 18:48:11 llama.go:408: waiting for llama runner to start responding
2023/10/22 18:48:11 llama.go:373: error starting llama runner: llama runner process has terminated
[GIN] 2023/10/22 - 18:48:11 | 500 | 4.608264866s | 127.0.0.1 | POST "/api/generate"
2023/10/22 18:48:11 llama.go:438: llama runner stopped with error: signal: illegal instruction (core dumped)
root@rig:/home/ubuntu#

Intel(R) Pentium(R) Gold G6400 CPU @ 4.00GHz
Description: Ubuntu 20.04.6 LTS

<!-- gh-comment-id:1774170144 --> @fabceolin commented on GitHub (Oct 22, 2023): same here: 2023/10/22 18:48:11 llama.go:356: starting llama runner 2023/10/22 18:48:11 llama.go:408: waiting for llama runner to start responding 2023/10/22 18:48:11 llama.go:373: error starting llama runner: llama runner process has terminated 2023/10/22 18:48:11 llama.go:438: llama runner stopped with error: signal: illegal instruction (core dumped) 2023/10/22 18:48:11 llama.go:356: starting llama runner 2023/10/22 18:48:11 llama.go:408: waiting for llama runner to start responding 2023/10/22 18:48:11 llama.go:373: error starting llama runner: llama runner process has terminated [GIN] 2023/10/22 - 18:48:11 | 500 | 4.608264866s | 127.0.0.1 | POST "/api/generate" 2023/10/22 18:48:11 llama.go:438: llama runner stopped with error: signal: illegal instruction (core dumped) root@rig:/home/ubuntu# Intel(R) Pentium(R) Gold G6400 CPU @ 4.00GHz Description: Ubuntu 20.04.6 LTS
GiteaMirror commented 2026-05-03 08:58:18 -05:00
Author
Owner
Copy Link

@randywreed commented on GitHub (Oct 22, 2023):

I was able to get this to work.
In my docker, I cloned the repo using gh cli
then I installed pr #871
and recompiled and it worked. Here's my Dockerfile:

FROM nvidia/cuda:11.8.0-devel-ubuntu22.04

ARG TARGETARCH="amd64"
ARG GOFLAGS="'-ldflags=-w -s'"

WORKDIR /go/src/github.com/jmorganca/ollama

# Update package list and install necessary tools
RUN apt-get update && apt-get install -y git build-essential cmake

# Download and install necessary tools
RUN apt-get update && apt-get install -y curl

# Download the GitHub CLI repository GPG key
RUN curl -fsSL https://cli.github.com/packages/githubcli-archive-keyring.gpg -o /usr/share/keyrings/githubcli-archive-keyring.gpg

# Update permissions on the GPG key
RUN chmod go+r /usr/share/keyrings/githubcli-archive-keyring.gpg

# Add the GitHub CLI repository to the APT sources list
RUN echo "deb [arch=$(dpkg --print-architecture) signed-by=/usr/share/keyrings/githubcli-archive-keyring.gpg] https://cli.github.com/packages stable main" | tee /etc/apt/sources.list.d/github-cli.list > /dev/null

# Update the APT package index and install the GitHub CLI
RUN apt-get update && apt-get install -y gh
# Authenticate with GitHub (assuming a personal access token is used)
# Replace 'your_token_here' with your actual GitHub personal access token, make sure it has repo.read permissions
RUN echo 'ghp_github_token' | gh auth login --with-token

# Clone the repository (including submodules) and checkout the specified pull request
RUN gh repo clone jmorganca/ollama /go/src/github.com/jmorganca/ollama -- --recurse-submodules && \
    cd /go/src/github.com/jmorganca/ollama && \
    gh pr checkout 871

# Download and setup Go
ADD https://dl.google.com/go/go1.21.3.linux-$TARGETARCH.tar.gz /tmp/go1.21.3.tar.gz
RUN mkdir -p /usr/local && tar xz -C /usr/local </tmp/go1.21.3.tar.gz

# Build the application
ENV GOARCH=$TARGETARCH
ENV GOFLAGS=$GOFLAGS
RUN /usr/local/go/bin/go generate ./... \
    && /usr/local/go/bin/go build .

# Create a new stage for the final image
FROM ubuntu:22.04 

RUN apt-get update && apt-get install -y ca-certificates
COPY --from=0 /go/src/github.com/jmorganca/ollama/ollama /bin/ollama
EXPOSE 11434
ENV OLLAMA_HOST 0.0.0.0
ENTRYPOINT ["/bin/ollama"]
CMD ["serve"]
<!-- gh-comment-id:1774202330 --> @randywreed commented on GitHub (Oct 22, 2023): I was able to get this to work. In my docker, I cloned the repo using gh cli then I installed pr #871 and recompiled and it worked. Here's my Dockerfile: ``` FROM nvidia/cuda:11.8.0-devel-ubuntu22.04 ARG TARGETARCH="amd64" ARG GOFLAGS="'-ldflags=-w -s'" WORKDIR /go/src/github.com/jmorganca/ollama # Update package list and install necessary tools RUN apt-get update && apt-get install -y git build-essential cmake # Download and install necessary tools RUN apt-get update && apt-get install -y curl # Download the GitHub CLI repository GPG key RUN curl -fsSL https://cli.github.com/packages/githubcli-archive-keyring.gpg -o /usr/share/keyrings/githubcli-archive-keyring.gpg # Update permissions on the GPG key RUN chmod go+r /usr/share/keyrings/githubcli-archive-keyring.gpg # Add the GitHub CLI repository to the APT sources list RUN echo "deb [arch=$(dpkg --print-architecture) signed-by=/usr/share/keyrings/githubcli-archive-keyring.gpg] https://cli.github.com/packages stable main" | tee /etc/apt/sources.list.d/github-cli.list > /dev/null # Update the APT package index and install the GitHub CLI RUN apt-get update && apt-get install -y gh # Authenticate with GitHub (assuming a personal access token is used) # Replace 'your_token_here' with your actual GitHub personal access token, make sure it has repo.read permissions RUN echo 'ghp_github_token' | gh auth login --with-token # Clone the repository (including submodules) and checkout the specified pull request RUN gh repo clone jmorganca/ollama /go/src/github.com/jmorganca/ollama -- --recurse-submodules && \ cd /go/src/github.com/jmorganca/ollama && \ gh pr checkout 871 # Download and setup Go ADD https://dl.google.com/go/go1.21.3.linux-$TARGETARCH.tar.gz /tmp/go1.21.3.tar.gz RUN mkdir -p /usr/local && tar xz -C /usr/local </tmp/go1.21.3.tar.gz # Build the application ENV GOARCH=$TARGETARCH ENV GOFLAGS=$GOFLAGS RUN /usr/local/go/bin/go generate ./... \ && /usr/local/go/bin/go build . # Create a new stage for the final image FROM ubuntu:22.04 RUN apt-get update && apt-get install -y ca-certificates COPY --from=0 /go/src/github.com/jmorganca/ollama/ollama /bin/ollama EXPOSE 11434 ENV OLLAMA_HOST 0.0.0.0 ENTRYPOINT ["/bin/ollama"] CMD ["serve"] ```
GiteaMirror commented 2026-05-03 08:58:19 -05:00
Author
Owner
Copy Link

@mxyng commented on GitHub (Oct 25, 2023):

This is likely related to #644

<!-- gh-comment-id:1779888869 --> @mxyng commented on GitHub (Oct 25, 2023): This is likely related to #644
GiteaMirror commented 2026-05-03 08:58:21 -05:00
Author
Owner
Copy Link

@jmorganca commented on GitHub (Oct 30, 2023):

Hi all! Ollama will now run on a larger range of CPUs as of version 0.1.6 and later. Note: AVX is still required – this should work for all CPUs released after 2011. If this doesn't fix the issue please feel free to re-open

<!-- gh-comment-id:1786123260 --> @jmorganca commented on GitHub (Oct 30, 2023): Hi all! Ollama will now run on a larger range of CPUs as of version `0.1.6` and later. Note: AVX is still required – this should work for all CPUs released after 2011. If this doesn't fix the issue please feel free to re-open
GiteaMirror commented 2026-05-03 08:58:22 -05:00
Author
Owner
Copy Link

@eviathan commented on GitHub (Dec 18, 2023):

I am still getting this issue I guess the computer I am trying to run it on is too old.

<!-- gh-comment-id:1861773197 --> @eviathan commented on GitHub (Dec 18, 2023): I am still getting this issue I guess the computer I am trying to run it on is too old.
GiteaMirror commented 2026-05-03 08:58:23 -05:00
Author
Owner
Copy Link

@Zetto12 commented on GitHub (Dec 19, 2023):

I got the same issue. Pretty old hardware too.

Running a Ubuntu Server VM in Proxmox on an HP Z800 with DDR3 and X5690 CPU.

<!-- gh-comment-id:1863550537 --> @Zetto12 commented on GitHub (Dec 19, 2023): I got the same issue. Pretty old hardware too. Running a Ubuntu Server VM in Proxmox on an HP Z800 with DDR3 and X5690 CPU.
GiteaMirror commented 2026-05-03 08:58:24 -05:00
Author
Owner
Copy Link

@Professor-Codephreak commented on GitHub (Dec 19, 2023):

Same error. Using Mint 9.1

verifying sha256 digest
writing manifest
removing any unused layers
success
Error: llama runner process has terminated

<!-- gh-comment-id:1863600020 --> @Professor-Codephreak commented on GitHub (Dec 19, 2023): Same error. Using Mint 9.1 verifying sha256 digest writing manifest removing any unused layers success Error: llama runner process has terminated
GiteaMirror commented 2026-05-03 08:58:25 -05:00
Author
Owner
Copy Link

@trifoil commented on GitHub (Dec 20, 2023):

Error: llama runner process has terminated

Running in a debian VM on TrueNAS
Intel(R) Xeon(R) CPU E5-2670 v2 @ 2.50GHz
DELL Poweredge R720

<!-- gh-comment-id:1864886783 --> @trifoil commented on GitHub (Dec 20, 2023): Error: llama runner process has terminated Running in a debian VM on TrueNAS Intel(R) Xeon(R) CPU E5-2670 v2 @ 2.50GHz DELL Poweredge R720
GiteaMirror commented 2026-05-03 08:58:26 -05:00
Author
Owner
Copy Link

@chrisdromey commented on GitHub (Dec 20, 2023):

Same issue.

Running in Ubuntu 22 VM on proxmox
Intel(R) Xeon(R) CPU E5-2620 v3 @ 2.40GHz


Dec 20 19:59:44 workstation ollama[756]: 2023/12/20 19:59:44 llama.go:436: starting llama runner
Dec 20 19:59:44 workstation ollama[756]: 2023/12/20 19:59:44 llama.go:494: waiting for llama runner to start responding
Dec 20 19:59:44 workstation ollama[756]: 2023/12/20 19:59:44 llama.go:451: signal: illegal instruction (core dumped)
Dec 20 19:59:44 workstation ollama[756]: 2023/12/20 19:59:44 llama.go:459: error starting llama runner: llama runner process has terminated
<!-- gh-comment-id:1865120371 --> @chrisdromey commented on GitHub (Dec 20, 2023): Same issue. Running in Ubuntu 22 VM on proxmox Intel(R) Xeon(R) CPU E5-2620 v3 @ 2.40GHz ``` Dec 20 19:59:44 workstation ollama[756]: 2023/12/20 19:59:44 llama.go:436: starting llama runner Dec 20 19:59:44 workstation ollama[756]: 2023/12/20 19:59:44 llama.go:494: waiting for llama runner to start responding Dec 20 19:59:44 workstation ollama[756]: 2023/12/20 19:59:44 llama.go:451: signal: illegal instruction (core dumped) Dec 20 19:59:44 workstation ollama[756]: 2023/12/20 19:59:44 llama.go:459: error starting llama runner: llama runner process has terminated ```
GiteaMirror commented 2026-05-03 08:58:26 -05:00
Author
Owner
Copy Link

@dosuken123 commented on GitHub (Dec 21, 2023):

Same issue.

MBA M2
OS: 14.0 (23A344)

Shinyas-Air:~ shinya$ ollama run llama2
pulling manifest 
pulling 22f7f8ef5f4c... 100% ▕████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▏ 3.8 GB                         
pulling 8c17c2ebb0ea... 100% ▕████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▏ 7.0 KB                         
pulling 7c23fb36d801... 100% ▕████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▏ 4.8 KB                         
pulling 2e0493f67d0c... 100% ▕████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▏   59 B                         
pulling 2759286baa87... 100% ▕████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▏  105 B                         
pulling 5407e3188df9... 100% ▕████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▏  529 B                         
verifying sha256 digest 
writing manifest 
removing any unused layers 
success 
Error: llama runner process has terminated
<!-- gh-comment-id:1865309591 --> @dosuken123 commented on GitHub (Dec 21, 2023): Same issue. MBA M2 OS: 14.0 (23A344) ``` Shinyas-Air:~ shinya$ ollama run llama2 pulling manifest pulling 22f7f8ef5f4c... 100% ▕████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▏ 3.8 GB pulling 8c17c2ebb0ea... 100% ▕████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▏ 7.0 KB pulling 7c23fb36d801... 100% ▕████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▏ 4.8 KB pulling 2e0493f67d0c... 100% ▕████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▏ 59 B pulling 2759286baa87... 100% ▕████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▏ 105 B pulling 5407e3188df9... 100% ▕████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▏ 529 B verifying sha256 digest writing manifest removing any unused layers success Error: llama runner process has terminated ```
GiteaMirror commented 2026-05-03 08:58:27 -05:00
Author
Owner
Copy Link

@dosuken123 commented on GitHub (Dec 21, 2023):

Ah, it turned out it's lacking memory https://github.com/jmorganca/ollama/issues/1508

<!-- gh-comment-id:1865315729 --> @dosuken123 commented on GitHub (Dec 21, 2023): Ah, it turned out it's lacking memory https://github.com/jmorganca/ollama/issues/1508
GiteaMirror commented 2026-05-03 08:58:28 -05:00
Author
Owner
Copy Link

@thommue commented on GitHub (Dec 26, 2023):

Same issue here:

Error: llama runner process has terminated

VM on Proxmox, with 25 cores and 64GB of Ram
The CPU is an AMD 5950X.

<!-- gh-comment-id:1869413571 --> @thommue commented on GitHub (Dec 26, 2023): Same issue here: Error: llama runner process has terminated VM on Proxmox, with 25 cores and 64GB of Ram The CPU is an AMD 5950X.
GiteaMirror commented 2026-05-03 08:58:28 -05:00
Author
Owner
Copy Link

@Bottlecap202 commented on GitHub (Dec 26, 2023):

Same issue here as well.
Error: llama runner process has terminated
Lenovo thinkstation c20
I was attempting to run with cuda but this was the most relevant

<!-- gh-comment-id:1869815577 --> @Bottlecap202 commented on GitHub (Dec 26, 2023): Same issue here as well. Error: llama runner process has terminated Lenovo thinkstation c20 I was attempting to run with cuda but this was the most relevant
GiteaMirror commented 2026-05-03 08:58:29 -05:00
Author
Owner
Copy Link

@timozn commented on GitHub (Dec 28, 2023):

Same error running in Docker on a QNAP TS-855eU with a Intel Atom C5125, 64GB RAM and a nVidia Quadro P1000:

docker exec -it ollama-ollama-1 ollama run llama2
Error: llama runner process has terminated

docker logs ollama-ollama-1                      
2023/12/28 22:36:46 images.go:737: total blobs: 6
2023/12/28 22:36:46 images.go:744: total unused blobs removed: 0
2023/12/28 22:36:46 routes.go:895: Listening on [::]:11434 (version 0.1.17)
[GIN] 2023/12/28 - 22:36:53 | 200 |      45.607µs |       127.0.0.1 | HEAD     "/"
[GIN] 2023/12/28 - 22:36:53 | 200 |    3.464403ms |       127.0.0.1 | POST     "/api/show"
[GIN] 2023/12/28 - 22:36:53 | 200 |    1.963032ms |       127.0.0.1 | POST     "/api/show"
{{false 2048 512 0 -1 0 false true false false true false true 0 0 0} 0 -1 -1 40 0.9 1 1 64 0.8 1.1 0 0 0 5 0.1 true [[INST] [/INST] <<SYS>> <</SYS>>]} 0 0
[--model /root/.ollama/models/blobs/sha256:8934d96d3f08982e95922b2b7a2c626a1fe873d7c3b06e8e56d7bc0a1fef9246 --ctx-size 2048 --batch-size 512 --n-gpu-layers 26 --embedding]
2023/12/28 22:36:56 llama.go:300: 4037 MB VRAM available, loading up to 26 GPU layers
2023/12/28 22:36:56 llama.go:440: starting llama runner
2023/12/28 22:36:56 llama.go:498: waiting for llama runner to start responding
2023/12/28 22:36:56 llama.go:455: signal: illegal instruction
2023/12/28 22:36:56 llama.go:463: error starting llama runner: llama runner process has terminated
2023/12/28 22:36:56 llama.go:529: llama runner stopped successfully
2023/12/28 22:36:56 llama.go:440: starting llama runner
2023/12/28 22:36:56 llama.go:498: waiting for llama runner to start responding
2023/12/28 22:36:56 llama.go:455: signal: illegal instruction
2023/12/28 22:36:56 llama.go:463: error starting llama runner: llama runner process has terminated
2023/12/28 22:36:56 llama.go:529: llama runner stopped successfully
[GIN] 2023/12/28 - 22:36:56 | 500 |  2.808293837s |       127.0.0.1 | POST     "/api/generate"

When running without GPU attached to the container, the logs look like this:

docker logs ollama-ollama-1                      
2023/12/28 22:44:32 images.go:737: total blobs: 6
2023/12/28 22:44:32 images.go:744: total unused blobs removed: 0
2023/12/28 22:44:32 routes.go:895: Listening on [::]:11434 (version 0.1.17)
2023/12/28 22:44:32 routes.go:915: warning: gpu support may not be enabled, check that you have installed GPU drivers: nvidia-smi command failed
[GIN] 2023/12/28 - 22:45:15 | 200 |      76.164µs |       127.0.0.1 | HEAD     "/"
[GIN] 2023/12/28 - 22:45:15 | 200 |     729.665µs |       127.0.0.1 | POST     "/api/show"
[GIN] 2023/12/28 - 22:45:15 | 200 |     547.037µs |       127.0.0.1 | POST     "/api/show"
{{false 2048 512 0 -1 0 false true false false true false true 0 0 0} 0 -1 -1 40 0.9 1 1 64 0.8 1.1 0 0 0 5 0.1 true [[INST] [/INST] <<SYS>> <</SYS>>]} 0 0
[--model /root/.ollama/models/blobs/sha256:8934d96d3f08982e95922b2b7a2c626a1fe873d7c3b06e8e56d7bc0a1fef9246 --ctx-size 2048 --batch-size 512 --n-gpu-layers 0 --embedding]
2023/12/28 22:45:16 llama.go:407: skipping accelerated runner because num_gpu=0
2023/12/28 22:45:16 llama.go:440: starting llama runner
2023/12/28 22:45:16 llama.go:498: waiting for llama runner to start responding
2023/12/28 22:45:16 llama.go:455: signal: illegal instruction
2023/12/28 22:45:16 llama.go:463: error starting llama runner: llama runner process has terminated
2023/12/28 22:45:16 llama.go:529: llama runner stopped successfully
[GIN] 2023/12/28 - 22:45:16 | 500 |  1.074327638s |       127.0.0.1 | POST     "/api/generate"

Edit: Just found out my CPU does not support AVX...

<!-- gh-comment-id:1871608764 --> @timozn commented on GitHub (Dec 28, 2023): Same error running in Docker on a QNAP TS-855eU with a Intel Atom C5125, 64GB RAM and a nVidia Quadro P1000: ``` docker exec -it ollama-ollama-1 ollama run llama2 Error: llama runner process has terminated ``` ``` docker logs ollama-ollama-1 2023/12/28 22:36:46 images.go:737: total blobs: 6 2023/12/28 22:36:46 images.go:744: total unused blobs removed: 0 2023/12/28 22:36:46 routes.go:895: Listening on [::]:11434 (version 0.1.17) [GIN] 2023/12/28 - 22:36:53 | 200 | 45.607µs | 127.0.0.1 | HEAD "/" [GIN] 2023/12/28 - 22:36:53 | 200 | 3.464403ms | 127.0.0.1 | POST "/api/show" [GIN] 2023/12/28 - 22:36:53 | 200 | 1.963032ms | 127.0.0.1 | POST "/api/show" {{false 2048 512 0 -1 0 false true false false true false true 0 0 0} 0 -1 -1 40 0.9 1 1 64 0.8 1.1 0 0 0 5 0.1 true [[INST] [/INST] <<SYS>> <</SYS>>]} 0 0 [--model /root/.ollama/models/blobs/sha256:8934d96d3f08982e95922b2b7a2c626a1fe873d7c3b06e8e56d7bc0a1fef9246 --ctx-size 2048 --batch-size 512 --n-gpu-layers 26 --embedding] 2023/12/28 22:36:56 llama.go:300: 4037 MB VRAM available, loading up to 26 GPU layers 2023/12/28 22:36:56 llama.go:440: starting llama runner 2023/12/28 22:36:56 llama.go:498: waiting for llama runner to start responding 2023/12/28 22:36:56 llama.go:455: signal: illegal instruction 2023/12/28 22:36:56 llama.go:463: error starting llama runner: llama runner process has terminated 2023/12/28 22:36:56 llama.go:529: llama runner stopped successfully 2023/12/28 22:36:56 llama.go:440: starting llama runner 2023/12/28 22:36:56 llama.go:498: waiting for llama runner to start responding 2023/12/28 22:36:56 llama.go:455: signal: illegal instruction 2023/12/28 22:36:56 llama.go:463: error starting llama runner: llama runner process has terminated 2023/12/28 22:36:56 llama.go:529: llama runner stopped successfully [GIN] 2023/12/28 - 22:36:56 | 500 | 2.808293837s | 127.0.0.1 | POST "/api/generate" ``` When running without GPU attached to the container, the logs look like this: ``` docker logs ollama-ollama-1 2023/12/28 22:44:32 images.go:737: total blobs: 6 2023/12/28 22:44:32 images.go:744: total unused blobs removed: 0 2023/12/28 22:44:32 routes.go:895: Listening on [::]:11434 (version 0.1.17) 2023/12/28 22:44:32 routes.go:915: warning: gpu support may not be enabled, check that you have installed GPU drivers: nvidia-smi command failed [GIN] 2023/12/28 - 22:45:15 | 200 | 76.164µs | 127.0.0.1 | HEAD "/" [GIN] 2023/12/28 - 22:45:15 | 200 | 729.665µs | 127.0.0.1 | POST "/api/show" [GIN] 2023/12/28 - 22:45:15 | 200 | 547.037µs | 127.0.0.1 | POST "/api/show" {{false 2048 512 0 -1 0 false true false false true false true 0 0 0} 0 -1 -1 40 0.9 1 1 64 0.8 1.1 0 0 0 5 0.1 true [[INST] [/INST] <<SYS>> <</SYS>>]} 0 0 [--model /root/.ollama/models/blobs/sha256:8934d96d3f08982e95922b2b7a2c626a1fe873d7c3b06e8e56d7bc0a1fef9246 --ctx-size 2048 --batch-size 512 --n-gpu-layers 0 --embedding] 2023/12/28 22:45:16 llama.go:407: skipping accelerated runner because num_gpu=0 2023/12/28 22:45:16 llama.go:440: starting llama runner 2023/12/28 22:45:16 llama.go:498: waiting for llama runner to start responding 2023/12/28 22:45:16 llama.go:455: signal: illegal instruction 2023/12/28 22:45:16 llama.go:463: error starting llama runner: llama runner process has terminated 2023/12/28 22:45:16 llama.go:529: llama runner stopped successfully [GIN] 2023/12/28 - 22:45:16 | 500 | 1.074327638s | 127.0.0.1 | POST "/api/generate" ``` Edit: Just found out my CPU does not support AVX...
GiteaMirror commented 2026-05-03 08:58:30 -05:00
Author
Owner
Copy Link

@mr-biz commented on GitHub (Dec 30, 2023):

Same issue here:

Error: llama runner process has terminated

VM on Proxmox, with 25 cores and 64GB of Ram The CPU is an AMD 5950X.

I encountered the same error on Proxmox. I resolved it by changing the VM CPU type from KVM64 (default) to host.
image

<!-- gh-comment-id:1872531426 --> @mr-biz commented on GitHub (Dec 30, 2023): > Same issue here: > > Error: llama runner process has terminated > > VM on Proxmox, with 25 cores and 64GB of Ram The CPU is an AMD 5950X. I encountered the same error on Proxmox. I resolved it by changing the VM CPU type from KVM64 (default) to host. <img width="589" alt="image" src="https://github.com/jmorganca/ollama/assets/44796116/fbd25c96-e98e-454d-a723-addc7dc1dbf0">
GiteaMirror commented 2026-05-03 08:58:32 -05:00
Author
Owner
Copy Link

@thommue commented on GitHub (Dec 30, 2023):

Same issue here:
Error: llama runner process has terminated
VM on Proxmox, with 25 cores and 64GB of Ram The CPU is an AMD 5950X.

I encountered the same error on Proxmox. I resolved it by changing the VM CPU type from KVM64 (default) to host. image

Yes it works!
Thanks for the help!

<!-- gh-comment-id:1872584955 --> @thommue commented on GitHub (Dec 30, 2023): > > Same issue here: > > Error: llama runner process has terminated > > VM on Proxmox, with 25 cores and 64GB of Ram The CPU is an AMD 5950X. > > I encountered the same error on Proxmox. I resolved it by changing the VM CPU type from KVM64 (default) to host. <img alt="image" width="589" src="https://private-user-images.githubusercontent.com/44796116/293463494-fbd25c96-e98e-454d-a723-addc7dc1dbf0.png?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.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.zyOvysIfZ3FUwZYoibKMwi_S3J0bTRhx2b5nXEjUnQs"> Yes it works! Thanks for the help!
GiteaMirror commented 2026-05-03 08:58:34 -05:00
Author
Owner
Copy Link

@kyuumeitai commented on GitHub (Jan 2, 2024):

Same issue here:
Error: llama runner process has terminated
VM on Proxmox, with 25 cores and 64GB of Ram The CPU is an AMD 5950X.

I encountered the same error on Proxmox. I resolved it by changing the VM CPU type from KVM64 (default) to host. image

thank you so much. For me in proxmox changing the vm CPU to host solved it.

<!-- gh-comment-id:1873590298 --> @kyuumeitai commented on GitHub (Jan 2, 2024): > > Same issue here: > > Error: llama runner process has terminated > > VM on Proxmox, with 25 cores and 64GB of Ram The CPU is an AMD 5950X. > > I encountered the same error on Proxmox. I resolved it by changing the VM CPU type from KVM64 (default) to host. <img alt="image" width="589" src="https://private-user-images.githubusercontent.com/44796116/293463494-fbd25c96-e98e-454d-a723-addc7dc1dbf0.png?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.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.XT7L5D6mAWmZ5-_buAk7_i10_eT6vCkFgDZfymqndPI"> thank you so much. For me in proxmox changing the vm CPU to host solved it.
GiteaMirror commented 2026-05-03 08:58:37 -05:00
Author
Owner
Copy Link

@idem2lyon commented on GitHub (Jan 9, 2024):

Hi all, is it possible to get the same option in esxi ?

<!-- gh-comment-id:1882535018 --> @idem2lyon commented on GitHub (Jan 9, 2024): Hi all, is it possible to get the same option in esxi ?
GiteaMirror commented 2026-05-03 08:58:39 -05:00
Author
Owner
Copy Link

@xdfnet commented on GitHub (May 14, 2024):

Same issue here:
ollama create eduaigc -f modelfile transferring model data using existing layer sha256:265610b1c6f4ad9d6d1d6377b3ebe39d348a3db5bc5d37185de7d554e31e6f75 creating new layer sha256:1507233b03785311a99a1beaf27f72d5a1c15abb3d2deba635defb2c68059262 creating new layer sha256:1da117d6723df114af0d948b614cae0aa684875e2775ca9607d23e2e0769651d creating new layer sha256:fa304d6750612c207b8705aca35391761f29492534e90b30575e4980d6ca82f6 creating new layer sha256:74454e71de083d29904e44fd8f5a7a0e5c1c020cd2553a0860a72bc4d5a9c6ef writing manifest success
ollama run eduaigc Error: llama runner process has terminated: exit status 0xc0000409

<!-- gh-comment-id:2110759766 --> @xdfnet commented on GitHub (May 14, 2024): Same issue here: `ollama create eduaigc -f modelfile transferring model data using existing layer sha256:265610b1c6f4ad9d6d1d6377b3ebe39d348a3db5bc5d37185de7d554e31e6f75 creating new layer sha256:1507233b03785311a99a1beaf27f72d5a1c15abb3d2deba635defb2c68059262 creating new layer sha256:1da117d6723df114af0d948b614cae0aa684875e2775ca9607d23e2e0769651d creating new layer sha256:fa304d6750612c207b8705aca35391761f29492534e90b30575e4980d6ca82f6 creating new layer sha256:74454e71de083d29904e44fd8f5a7a0e5c1c020cd2553a0860a72bc4d5a9c6ef writing manifest success` ` ollama run eduaigc Error: llama runner process has terminated: exit status 0xc0000409`
GiteaMirror commented 2026-05-03 08:58:39 -05:00
Author
Owner
Copy Link

@vishalharkal15 commented on GitHub (Jun 27, 2024):

https://github.com/login?return_to=https%3A%2F%2Fgithub.com%2Follama%2Follama%2Fissues%2F845
Uploading Screenshot 2024-06-27 190147.png…

<!-- gh-comment-id:2194714283 --> @vishalharkal15 commented on GitHub (Jun 27, 2024): https://github.com/login?return_to=https%3A%2F%2Fgithub.com%2Follama%2Follama%2Fissues%2F845 ![Uploading Screenshot 2024-06-27 190147.png…]()
GiteaMirror commented 2026-05-03 08:58:41 -05:00
Author
Owner
Copy Link

@vishalharkal15 commented on GitHub (Jun 27, 2024):

Screenshot 2024-06-27 185319
Screenshot 2024-06-27 184756

<!-- gh-comment-id:2194714424 --> @vishalharkal15 commented on GitHub (Jun 27, 2024): ![Screenshot 2024-06-27 185319](https://github.com/ollama/ollama/assets/168349626/e7b4e0d9-d032-4e7b-b26e-f71b4969f6a4) ![Screenshot 2024-06-27 184756](https://github.com/ollama/ollama/assets/168349626/9d271da4-647d-49b6-bfd3-92c78bbd2e75)
GiteaMirror commented 2026-05-03 08:58:41 -05:00
Author
Owner
Copy Link

@121212wwd commented on GitHub (Jul 7, 2024):

i just have an issue when downloading the codegemma 7b, i am going to try install the lesser version, but i have 4gb or ram and some really 4SS CPU, i'm thinking maybe its not enough but seeing that there was a guy that had some good specs including a GPU, maybe the computers just aren't enough to handle these ai apis. my sinopsis or conclusion is that maybe the type of hardware is needed to handle the AI's we just dont have, maybe one person had an AMD GPU but it says it looks for a NVIDIA gpu, if anyone has something like an NVIDIA gpu, or some specs that can handle this sort of thing post down, i'm very curious about this.

oh and i think i read half of this thread XD

<!-- gh-comment-id:2212456205 --> @121212wwd commented on GitHub (Jul 7, 2024): i just have an issue when downloading the codegemma 7b, i am going to try install the lesser version, but i have 4gb or ram and some really 4SS CPU, i'm thinking maybe its not enough but seeing that there was a guy that had some good specs including a GPU, maybe the computers just aren't enough to handle these ai apis. my sinopsis or conclusion is that maybe the type of hardware is needed to handle the AI's we just dont have, maybe one person had an AMD GPU but it says it looks for a NVIDIA gpu, if anyone has something like an NVIDIA gpu, or some specs that can handle this sort of thing post down, i'm very curious about this. oh and i think i read half of this thread XD
GiteaMirror commented 2026-05-03 08:58:42 -05:00
Author
Owner
Copy Link

@vishalharkal15 commented on GitHub (Jul 12, 2024):

I've encountered an issue while downloading the CodeGemma 7B model. Before trying to install a lesser version, I wanted to share some details about my hardware. I have 4GB of RAM and a 4SS CPU. I'm starting to think that my setup might not be sufficient to handle these AI APIs.

I noticed a post from someone with good specs, including a GPU, who was also facing issues. This makes me wonder if our computers might just not be powerful enough to handle these AI models.

Additionally, it seems that CodeGemma is looking for an NVIDIA GPU, but one person mentioned having an AMD GPU. If anyone has an NVIDIA GPU or hardware that can handle these kinds of models, could you please share your specs? I'm very curious to see what kind of hardware is needed to run this smoothly.

<!-- gh-comment-id:2225937861 --> @vishalharkal15 commented on GitHub (Jul 12, 2024): I've encountered an issue while downloading the CodeGemma 7B model. Before trying to install a lesser version, I wanted to share some details about my hardware. I have 4GB of RAM and a 4SS CPU. I'm starting to think that my setup might not be sufficient to handle these AI APIs. I noticed a post from someone with good specs, including a GPU, who was also facing issues. This makes me wonder if our computers might just not be powerful enough to handle these AI models. Additionally, it seems that CodeGemma is looking for an NVIDIA GPU, but one person mentioned having an AMD GPU. If anyone has an NVIDIA GPU or hardware that can handle these kinds of models, could you please share your specs? I'm very curious to see what kind of hardware is needed to run this smoothly.
GiteaMirror commented 2026-05-03 08:58:43 -05:00
Author
Owner
Copy Link

@121212wwd commented on GitHub (Jul 12, 2024):

dude, did you just steal my comment and post it as your own? actually post something of your own

<!-- gh-comment-id:2225942145 --> @121212wwd commented on GitHub (Jul 12, 2024): dude, did you just steal my comment and post it as your own? actually post something of your own
GiteaMirror commented 2026-05-03 08:58:44 -05:00
Author
Owner
Copy Link

@Bhavya031 commented on GitHub (Jul 21, 2024):

dude, did you just steal my comment and post it as your own? actually post something of your own

Yeah, I think he is trying to comment to get a contribution highlight on his GitHub page. He also posted a screenshot of installing Ollama on Windows, which is irrelevant to this topic. Above that, there is some other screenshot link. I think he doesn't understand that you have to make a commit to get that dot.

<!-- gh-comment-id:2241529877 --> @Bhavya031 commented on GitHub (Jul 21, 2024): > dude, did you just steal my comment and post it as your own? actually post something of your own Yeah, I think he is trying to comment to get a contribution highlight on his GitHub page. He also posted a screenshot of installing Ollama on Windows, which is irrelevant to this topic. Above that, there is some other screenshot link. I think he doesn't understand that you have to make a commit to get that dot.
GiteaMirror commented 2026-05-03 08:58:44 -05:00
Author
Owner
Copy Link

@121212wwd commented on GitHub (Jul 29, 2024):

Add your specs in a new comment, maybe it's because you haven't got the power to run it or maybe try and find an answer in the thread, it's closed so someone has already found the problem and fixed it.

Good Luck

DOODES MALOODES OUT >>

<!-- gh-comment-id:2254733783 --> @121212wwd commented on GitHub (Jul 29, 2024): Add your specs in a new comment, maybe it's because you haven't got the power to run it or maybe try and find an answer in the thread, it's closed so someone has already found the problem and fixed it. Good Luck DOODES MALOODES OUT >>
GiteaMirror commented 2026-05-03 08:58:45 -05:00
Author
Owner
Copy Link

@121212wwd commented on GitHub (Jul 29, 2024):

ITS A MEMORY PROBLEM, IF NOT THEN GOD KNOWS WHAT IT IS

<!-- gh-comment-id:2254734584 --> @121212wwd commented on GitHub (Jul 29, 2024): ITS A MEMORY PROBLEM, IF NOT THEN GOD KNOWS WHAT IT IS
GiteaMirror commented 2026-05-03 08:58:46 -05:00
Author
Owner
Copy Link

@121212wwd commented on GitHub (Aug 3, 2024):

@ISMAEL45966 PLEASE READ THE ABOVE COMMENT, Your lack of observation is concerning. If you are running an AI it is very power and resource dependent. SO if you have a laptop it's not the best idea to run Deepseek code VERSION 2 ON A LAPTOP. Please read this and stop commenting, i have an answer for you, READ IT

<!-- gh-comment-id:2266320855 --> @121212wwd commented on GitHub (Aug 3, 2024): @ISMAEL45966 PLEASE READ THE ABOVE COMMENT, Your lack of observation is concerning. If you are running an AI it is very power and resource dependent. SO if you have a laptop it's not the best idea to run Deepseek code VERSION 2 ON A LAPTOP. Please read this and stop commenting, i have an answer for you, READ IT
GiteaMirror commented 2026-05-03 08:58:46 -05:00
Author
Owner
Copy Link

@RobRoyce commented on GitHub (Aug 30, 2024):

I am getting the same error while trying to run llama3.1:8b (~4GB) on a 64GB M2 Mac. The same model ran two days ago just fine. This is not a memory problem.

<!-- gh-comment-id:2322098484 --> @RobRoyce commented on GitHub (Aug 30, 2024): I am getting the same error while trying to run `llama3.1:8b` (~4GB) on a 64GB M2 Mac. The same model ran two days ago just fine. This is _not_ a memory problem.
GiteaMirror commented 2026-05-03 08:58:47 -05:00
Author
Owner
Copy Link

@khuezy commented on GitHub (Aug 30, 2024):

@RobRoyce Hey I have the same exact problem. It was working yesterday but now my ollama returns 500 Internal Server Error

<!-- gh-comment-id:2322111119 --> @khuezy commented on GitHub (Aug 30, 2024): @RobRoyce Hey I have the same exact problem. It was working yesterday but now my ollama returns 500 Internal Server Error
GiteaMirror commented 2026-05-03 08:58:47 -05:00
Author
Owner
Copy Link

@RobRoyce commented on GitHub (Aug 30, 2024):

@RobRoyce Hey I have the same exact problem. It was working yesterday but now my ollama returns 500 Internal Server Error

I was previously running in Docker. After installing Ollama locally, it magically started working. My Docker resources were set to provide 24GB of RAM, so I still don't see why it failed.

<!-- gh-comment-id:2322181312 --> @RobRoyce commented on GitHub (Aug 30, 2024): > @RobRoyce Hey I have the same exact problem. It was working yesterday but now my ollama returns 500 Internal Server Error I was previously running in Docker. After installing Ollama locally, it magically started working. My Docker resources were set to provide 24GB of RAM, so I still don't see why it failed.
GiteaMirror commented 2026-05-03 08:58:48 -05:00
Author
Owner
Copy Link

@khuezy commented on GitHub (Aug 30, 2024):

@RobRoyce I believe 0.3.8 (3 days ago) broke ollama in docker.
Try curl -fsSL https://ollama.com/install.sh | OLLAMA_VERSION=0.3.6 sh
It's always a good idea to lock down the specific version instead of running latest.

<!-- gh-comment-id:2322229123 --> @khuezy commented on GitHub (Aug 30, 2024): @RobRoyce I believe 0.3.8 (3 days ago) broke ollama in docker. Try `curl -fsSL https://ollama.com/install.sh | OLLAMA_VERSION=0.3.6 sh` It's always a good idea to lock down the specific version instead of running latest.
GiteaMirror commented 2026-05-03 08:58:49 -05:00
Author
Owner
Copy Link

@RobRoyce commented on GitHub (Aug 30, 2024):

Thank you, I'll give this a shot!

<!-- gh-comment-id:2322259296 --> @RobRoyce commented on GitHub (Aug 30, 2024): Thank you, I'll give this a shot!
GiteaMirror commented 2026-05-03 08:58:50 -05:00
Author
Owner
Copy Link

@121212wwd commented on GitHub (Aug 30, 2024):

Hey, i think this is a memory problem for people that have laptops and try to run something where you NEED that memory, but you've found the solution. so have fun.
maybe it's docker, maybe it's RAM, but my comment was just an assumption and warns people that use laptops with specs that can't handle ollama, to not use ollama, instead use chatgpt.
I hope you got your issue resolved but i only made an assumption while also being very positive and strong that that was the issue.
My bad.
I, in total, have 4gbs of memory and i'm running in linux. Bit of a different situation for me, if you have a solution plz comment, i'd really like to try on my pc instead.
got to try it first though.

<!-- gh-comment-id:2322312287 --> @121212wwd commented on GitHub (Aug 30, 2024): Hey, i think this is a memory problem for people that have laptops and try to run something where you NEED that memory, but you've found the solution. so have fun. maybe it's docker, maybe it's RAM, but my comment was just an assumption and warns people that use laptops with specs that can't handle ollama, to not use ollama, instead use chatgpt. I hope you got your issue resolved but i only made an assumption while also being very positive and strong that that was the issue. My bad. I, in total, have 4gbs of memory and i'm running in linux. Bit of a different situation for me, if you have a solution plz comment, i'd really like to try on my pc instead. got to try it first though.
GiteaMirror commented 2026-05-03 08:58:51 -05:00
Author
Owner
Copy Link

@ShivamSrng commented on GitHub (Nov 5, 2024):

I am getting the same error while trying to run llama3.1:8b (~4GB) on a 64GB M2 Mac. The same model ran two days ago just fine. This is not a memory problem.

Were you able to solve it ? I am facing the same issue as yours. Here's how it looks in case:
CUDA error: an illegal instruction was encountered
current device: 0, in function ggml_backend_cuda_synchronize at C:\a\ollama\ollama\llm\llama.cpp\ggml\src\ggml-cuda.cu:2473
cudaStreamSynchronize(cuda_ctx->stream())
C:\a\ollama\ollama\llm\llama.cpp\ggml\src\ggml-cuda.cu:106: CUDA error

OS: Windows
GPU: Nvidia
CPU: Intel
Ollama version: 0.3.14
Driver Version: 552.44
CUDA Version: 12.4
RAM: 32GB

<!-- gh-comment-id:2456309026 --> @ShivamSrng commented on GitHub (Nov 5, 2024): > I am getting the same error while trying to run `llama3.1:8b` (~4GB) on a 64GB M2 Mac. The same model ran two days ago just fine. This is _not_ a memory problem. Were you able to solve it ? I am facing the same issue as yours. Here's how it looks in case: CUDA error: an illegal instruction was encountered current device: 0, in function ggml_backend_cuda_synchronize at C:\a\ollama\ollama\llm\llama.cpp\ggml\src\ggml-cuda.cu:2473 cudaStreamSynchronize(cuda_ctx->stream()) C:\a\ollama\ollama\llm\llama.cpp\ggml\src\ggml-cuda.cu:106: CUDA error OS: Windows GPU: Nvidia CPU: Intel Ollama version: 0.3.14 Driver Version: 552.44 CUDA Version: 12.4 RAM: 32GB
GiteaMirror commented 2026-05-03 08:58:53 -05:00
Author
Owner
Copy Link

@121212wwd commented on GitHub (Nov 8, 2024):

That means an illegal thing was done, don't override it or break it. Try to reinstall. I also think with bigger models you might need a beefier pc, don't use a laptop it will melt

<!-- gh-comment-id:2464427450 --> @121212wwd commented on GitHub (Nov 8, 2024): That means an illegal thing was done, don't override it or break it. Try to reinstall. I also think with bigger models you might need a beefier pc, don't use a laptop it will melt
GiteaMirror commented 2026-05-03 08:58:57 -05:00
Author
Owner
Copy Link

@smdhussain06 commented on GitHub (Feb 5, 2025):

Same error using in android

<!-- gh-comment-id:2635849610 --> @smdhussain06 commented on GitHub (Feb 5, 2025): Same error using in android
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
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#62440
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?