[GH-ISSUE #2123] After upgrading Ollama. It just doesn't run anymore any model: Error: Post "http://127.0.0.1:11434/api/generate": EOF #1214

New Issue

GiteaMirror · 2026-04-12T10:59:19-05:00

GiteaMirror commented

2026-04-12 10:59:19 -05:00

Originally created by @venturaEffect on GitHub (Jan 21, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/2123

I see this error everywhere during months. There are plenty of releases but this error continuesly appears.

Followed on Discord, searched on the web, saw issues on the repo related to this.

Did things like creating a Modelfile "dolphin-mistral" with FROM dolphin-2.1-mistral-7b PARAMETER num_gpu 0.

Ollama upgraded: curl https://ollama.ai/install.sh | sh

Nothing:

ollama run dolphin-mistral Error: Post "http://127.0.0.1:11434/api/generate": EOF

Any suggestion will be much appreciated.

Originally created by @venturaEffect on GitHub (Jan 21, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/2123 I see this error everywhere during months. There are plenty of releases but this error continuesly appears. Followed on Discord, searched on the web, saw issues on the repo related to this. Did things like creating a Modelfile "dolphin-mistral" with `FROM dolphin-2.1-mistral-7b PARAMETER num_gpu 0`. Ollama upgraded: `curl https://ollama.ai/install.sh | sh` Nothing: `ollama run dolphin-mistral Error: Post "http://127.0.0.1:11434/api/generate": EOF` Any suggestion will be much appreciated.

GiteaMirror closed this issue

2026-04-12 10:59:19 -05:00

GiteaMirror commented

2026-04-12 10:59:20 -05:00

@ssebastianoo commented on GitHub (Jan 21, 2024):

Same thing, if I do ollama run llama2 it works fine but ollama run mario (created from this) raises this error:

Error: Post "http://127.0.0.1:11434/api/generate": EOF

@ssebastianoo commented on GitHub (Jan 21, 2024): Same thing, if I do `ollama run llama2` it works fine but `ollama run mario` (created from [this](https://github.com/jmorganca/ollama?tab=readme-ov-file#customize-a-prompt)) raises this error: ``` Error: Post "http://127.0.0.1:11434/api/generate": EOF ```

GiteaMirror commented

2026-04-12 10:59:20 -05:00

@ssebastianoo commented on GitHub (Jan 21, 2024):

watching ollama serve I found out this:

2024/01/21 16:26:56 images.go:430: [model] - llama2
2024/01/21 16:26:56 images.go:430: [temperature] - 1
2024/01/21 16:26:56 images.go:430: [system] - 
You are Mario from super mario bros, acting as an assistant.
[GIN] 2024/01/21 - 16:26:56 | 200 |    2.255856ms |       127.0.0.1 | POST     "/api/create"
[GIN] 2024/01/21 - 16:27:04 | 200 |      37.785µs |       127.0.0.1 | HEAD     "/"
[GIN] 2024/01/21 - 16:27:04 | 200 |     835.181µs |       127.0.0.1 | POST     "/api/show"
[GIN] 2024/01/21 - 16:27:04 | 200 |     741.592µs |       127.0.0.1 | POST     "/api/show"
[GIN] 2024/01/21 - 16:27:04 | 200 |     549.943µs |       127.0.0.1 | POST     "/api/generate"
2024/01/21 16:27:05 ext_server_common.go:158: loaded 0 images

CUDA error 2 at /go/src/github.com/jmorganca/ollama/llm/llama.cpp/ggml-cuda.cu:6600: out of memory
current device: 0
GGML_ASSERT: /go/src/github.com/jmorganca/ollama/llm/llama.cpp/ggml-cuda.cu:6600: !"CUDA error"
Aborted

@ssebastianoo commented on GitHub (Jan 21, 2024): watching `ollama serve` I found out this: ``` 2024/01/21 16:26:56 images.go:430: [model] - llama2 2024/01/21 16:26:56 images.go:430: [temperature] - 1 2024/01/21 16:26:56 images.go:430: [system] - You are Mario from super mario bros, acting as an assistant. [GIN] 2024/01/21 - 16:26:56 | 200 | 2.255856ms | 127.0.0.1 | POST "/api/create" [GIN] 2024/01/21 - 16:27:04 | 200 | 37.785µs | 127.0.0.1 | HEAD "/" [GIN] 2024/01/21 - 16:27:04 | 200 | 835.181µs | 127.0.0.1 | POST "/api/show" [GIN] 2024/01/21 - 16:27:04 | 200 | 741.592µs | 127.0.0.1 | POST "/api/show" [GIN] 2024/01/21 - 16:27:04 | 200 | 549.943µs | 127.0.0.1 | POST "/api/generate" 2024/01/21 16:27:05 ext_server_common.go:158: loaded 0 images CUDA error 2 at /go/src/github.com/jmorganca/ollama/llm/llama.cpp/ggml-cuda.cu:6600: out of memory current device: 0 GGML_ASSERT: /go/src/github.com/jmorganca/ollama/llm/llama.cpp/ggml-cuda.cu:6600: !"CUDA error" Aborted ```

GiteaMirror commented

2026-04-12 10:59:20 -05:00

@venturaEffect commented on GitHub (Jan 21, 2024):

Thanks for sharing. Did you solve it? It seems more people are facing this issue.

@venturaEffect commented on GitHub (Jan 21, 2024): Thanks for sharing. Did you solve it? It seems more people are facing this issue.

GiteaMirror commented

2026-04-12 10:59:21 -05:00

@basillicus commented on GitHub (Jan 21, 2024):

It may be related to #1952 ?

@basillicus commented on GitHub (Jan 21, 2024): It may be related to #1952 ?

GiteaMirror commented

2026-04-12 10:59:21 -05:00

@venturaEffect commented on GitHub (Jan 21, 2024):

It may be related to #1952 ?

It could and makes sense as I'm using it by doing a RAG on Langchain. But there is no really a workaround without the RAG. Is there any solution that you know that could solve the issue? I'm using dolphin-mistral because is a good one and needs to be uncensored.

Appreciate

@venturaEffect commented on GitHub (Jan 21, 2024): > It may be related to #1952 ? It could and makes sense as I'm using it by doing a RAG on Langchain. But there is no really a workaround without the RAG. Is there any solution that you know that could solve the issue? I'm using dolphin-mistral because is a good one and needs to be uncensored. Appreciate

GiteaMirror commented

2026-04-12 10:59:22 -05:00

@t0m3k commented on GitHub (Jan 21, 2024):

Same problem for me on Manjaro, 6900 xtx:

Ollama serve

> ollama serve
2024/01/21 22:00:11 images.go:810: INFO total blobs: 6
2024/01/21 22:00:11 images.go:817: INFO total unused blobs removed: 0
2024/01/21 22:00:11 routes.go:943: INFO Listening on 127.0.0.1:11434 (version 0.1.21)
2024/01/21 22:00:11 payload_common.go:106: INFO Extracting dynamic libraries...
2024/01/21 22:00:13 payload_common.go:145: INFO Dynamic LLM libraries [rocm_v6 cpu cpu_avx cuda_v11 rocm_v5 cpu_avx2]
2024/01/21 22:00:13 gpu.go:91: INFO Detecting GPU type
2024/01/21 22:00:13 gpu.go:210: INFO Searching for GPU management library libnvidia-ml.so
2024/01/21 22:00:13 gpu.go:256: INFO Discovered GPU libraries: []
2024/01/21 22:00:13 gpu.go:210: INFO Searching for GPU management library librocm_smi64.so
2024/01/21 22:00:13 gpu.go:256: INFO Discovered GPU libraries: [/opt/rocm/lib/librocm_smi64.so.5.0]
2024/01/21 22:00:13 gpu.go:106: INFO Radeon GPU detected
[GIN] 2024/01/21 - 22:00:15 | 200 |       40.73µs |       127.0.0.1 | HEAD     "/"
[GIN] 2024/01/21 - 22:00:15 | 200 |     376.902µs |       127.0.0.1 | POST     "/api/show"
[GIN] 2024/01/21 - 22:00:15 | 200 |     236.512µs |       127.0.0.1 | POST     "/api/show"
2024/01/21 22:00:15 cpu_common.go:11: INFO CPU has AVX2
loading library /tmp/ollama1546965028/rocm_v5/libext_server.so
2024/01/21 22:00:15 dyn_ext_server.go:90: INFO Loading Dynamic llm server: /tmp/ollama1546965028/rocm_v5/libext_server.so
2024/01/21 22:00:15 dyn_ext_server.go:139: INFO Initializing llama server
free(): invalid pointer
[1]    275518 IOT instruction (core dumped)  ollama serve

Run

❯ ollama run codellama
Error: Post "http://127.0.0.1:11434/api/chat": EOF

@t0m3k commented on GitHub (Jan 21, 2024): Same problem for me on Manjaro, 6900 xtx: ## Ollama serve ``` > ollama serve 2024/01/21 22:00:11 images.go:810: INFO total blobs: 6 2024/01/21 22:00:11 images.go:817: INFO total unused blobs removed: 0 2024/01/21 22:00:11 routes.go:943: INFO Listening on 127.0.0.1:11434 (version 0.1.21) 2024/01/21 22:00:11 payload_common.go:106: INFO Extracting dynamic libraries... 2024/01/21 22:00:13 payload_common.go:145: INFO Dynamic LLM libraries [rocm_v6 cpu cpu_avx cuda_v11 rocm_v5 cpu_avx2] 2024/01/21 22:00:13 gpu.go:91: INFO Detecting GPU type 2024/01/21 22:00:13 gpu.go:210: INFO Searching for GPU management library libnvidia-ml.so 2024/01/21 22:00:13 gpu.go:256: INFO Discovered GPU libraries: [] 2024/01/21 22:00:13 gpu.go:210: INFO Searching for GPU management library librocm_smi64.so 2024/01/21 22:00:13 gpu.go:256: INFO Discovered GPU libraries: [/opt/rocm/lib/librocm_smi64.so.5.0] 2024/01/21 22:00:13 gpu.go:106: INFO Radeon GPU detected [GIN] 2024/01/21 - 22:00:15 | 200 | 40.73µs | 127.0.0.1 | HEAD "/" [GIN] 2024/01/21 - 22:00:15 | 200 | 376.902µs | 127.0.0.1 | POST "/api/show" [GIN] 2024/01/21 - 22:00:15 | 200 | 236.512µs | 127.0.0.1 | POST "/api/show" 2024/01/21 22:00:15 cpu_common.go:11: INFO CPU has AVX2 loading library /tmp/ollama1546965028/rocm_v5/libext_server.so 2024/01/21 22:00:15 dyn_ext_server.go:90: INFO Loading Dynamic llm server: /tmp/ollama1546965028/rocm_v5/libext_server.so 2024/01/21 22:00:15 dyn_ext_server.go:139: INFO Initializing llama server free(): invalid pointer [1] 275518 IOT instruction (core dumped) ollama serve ``` ## Run ``` ❯ ollama run codellama Error: Post "http://127.0.0.1:11434/api/chat": EOF ```

GiteaMirror commented

2026-04-12 10:59:22 -05:00

@ssebastianoo commented on GitHub (Jan 21, 2024):

I can't rn but did anyone try to do the same things using an older version from the releases?

@ssebastianoo commented on GitHub (Jan 21, 2024): I can't rn but did anyone try to do the same things using an older version from the releases?

GiteaMirror commented

2026-04-12 10:59:23 -05:00

@dhiltgen commented on GitHub (Jan 22, 2024):

@venturaEffect could you provide the server logs so we can see why it crashed?

@ssebastianoo as others have noted, we're continuing to refine our memory prediction logic to balance using as much GPU memory as possible, without exceeding the capacity. Can you clarify which version of ollama you were running? 0.1.21 has fixes that may solve this for you, but if you still see OOMs please let us know.

@t0m3k your crash looks like a Radeon related defect. Depending on what @venturaEffect ran into, we might want to track the Radeon crash with a different issue.

@dhiltgen commented on GitHub (Jan 22, 2024): @venturaEffect could you provide the server logs so we can see why it crashed? @ssebastianoo as others have noted, we're continuing to refine our memory prediction logic to balance using as much GPU memory as possible, without exceeding the capacity. Can you clarify which version of ollama you were running? [0.1.21](https://github.com/jmorganca/ollama/releases/tag/v0.1.21) has fixes that may solve this for you, but if you still see OOMs please let us know. @t0m3k your crash looks like a Radeon related defect. Depending on what @venturaEffect ran into, we might want to track the Radeon crash with a different issue.

GiteaMirror commented

2026-04-12 10:59:23 -05:00

@venturaEffect commented on GitHub (Jan 22, 2024):

@dhiltgen thanks! I have downgraded to older version suggested by @jmorganca . This solved the issue but now I'm facing another problem and it is that I can't use it for a RAG on Langchain because the context window is very limited. So it isn't useful at all. I'm trying to figure out how to solve this but it seems with Ollama llms it looks like a no exit road. Don't like it because we are loosing the power to use all these llms and are depending again on OpenAI and it's polite GPT. If you know any solution would be super appreciated 👍

@venturaEffect commented on GitHub (Jan 22, 2024): @dhiltgen thanks! I have downgraded to older version suggested by @jmorganca . This solved the issue but now I'm facing another problem and it is that I can't use it for a RAG on Langchain because the context window is very limited. So it isn't useful at all. I'm trying to figure out how to solve this but it seems with Ollama llms it looks like a no exit road. Don't like it because we are loosing the power to use all these llms and are depending again on OpenAI and it's polite GPT. If you know any solution would be super appreciated 👍

GiteaMirror commented

2026-04-12 10:59:23 -05:00

@nathaniel-brough commented on GitHub (Jan 23, 2024):

I'm also experiencing this problem but only in some cases. e.g.

# These work fine
$ ollama run phi # 1.6GB, 2.7B parameters
$ ollama run llama2 # 3.8GB, 7B parameters

# This crashes with the same error
$ ollama run stable-code  # 1.6GB, 3B parameters
Error: Post "http://127.0.0.1:11434/api/generate": EOF

I have been running these using the docker image on a fairly low end laptop GPU and/or hybrid CPU in the case of llama2.

GPU specs:

NVIDIA GeForce MX150
CUDA core 384
Total dedicated memory 2048MB

I did find it interesting that stable-code 3B parameters, is approximately the same size as phi with 2.7B parameters. I would have expected the size to be about 10% difference between the two models. Perhaps there is some miscalculation in the model size which might make the CUDA memory estimation wrong?

@dhiltgen I've attached the server log here. i.e. the output of docker logs ollama 2> ~/ollama_crash.txt
ollama_crash.txt

@nathaniel-brough commented on GitHub (Jan 23, 2024): I'm also experiencing this problem but only in some cases. e.g. ``` # These work fine $ ollama run phi # 1.6GB, 2.7B parameters $ ollama run llama2 # 3.8GB, 7B parameters # This crashes with the same error $ ollama run stable-code # 1.6GB, 3B parameters Error: Post "http://127.0.0.1:11434/api/generate": EOF ``` I have been running these using the docker image on a fairly low end laptop GPU and/or hybrid CPU in the case of llama2. GPU specs: - NVIDIA GeForce MX150 - CUDA core 384 - Total dedicated memory 2048MB I did find it interesting that `stable-code` 3B parameters, is approximately the same size as `phi` with 2.7B parameters. I would have expected the size to be about 10% difference between the two models. Perhaps there is some miscalculation in the model size which might make the CUDA memory estimation wrong? @dhiltgen I've attached the server log here. i.e. the output of `docker logs ollama 2> ~/ollama_crash.txt` [ollama_crash.txt](https://github.com/jmorganca/ollama/files/14028441/ollama_crash.txt)

GiteaMirror commented

2026-04-12 10:59:24 -05:00

@dhiltgen commented on GitHub (Jan 23, 2024):

@silvergasp looks like you hit a GPU out-of-memory on 0.1.20. We've added some fixes to 0.1.21 to improve low memory GPUs, but the algorithm still isn't quite perfect.

@venturaEffect if you can try with 0.1.21 and share the server logs that will help us understand if this is a known issue we're working on, or something new.

@dhiltgen commented on GitHub (Jan 23, 2024): @silvergasp looks like you hit a GPU out-of-memory on 0.1.20. We've added some fixes to [0.1.21](https://github.com/ollama/ollama/releases/tag/v0.1.21) to improve low memory GPUs, but the algorithm still isn't quite perfect. @venturaEffect if you can try with [0.1.21](https://github.com/ollama/ollama/releases/tag/v0.1.21) and share the server logs that will help us understand if this is a known issue we're working on, or something new.

GiteaMirror commented

2026-04-12 10:59:24 -05:00

@YeryODell commented on GitHub (Jan 24, 2024):

@venturaEffect could you provide the server logs so we can see why it crashed?

@ssebastianoo as others have noted, we're continuing to refine our memory prediction logic to balance using as much GPU memory as possible, without exceeding the capacity. Can you clarify which version of ollama you were running? 0.1.21 has fixes that may solve this for you, but if you still see OOMs please let us know.

@t0m3k your crash looks like a Radeon related defect. Depending on what @venturaEffect ran into, we might want to track the Radeon crash with a different issue.

I'm having the same exact issue and have tried all the same fixes. I can't find any instructions on how to check my version or to upgrade to a pre-release version. Can you please provide instructions for Ubuntu?

@YeryODell commented on GitHub (Jan 24, 2024): > @venturaEffect could you provide the server logs so we can see why it crashed? > > @ssebastianoo as others have noted, we're continuing to refine our memory prediction logic to balance using as much GPU memory as possible, without exceeding the capacity. Can you clarify which version of ollama you were running? [0.1.21](https://github.com/jmorganca/ollama/releases/tag/v0.1.21) has fixes that may solve this for you, but if you still see OOMs please let us know. > > @t0m3k your crash looks like a Radeon related defect. Depending on what @venturaEffect ran into, we might want to track the Radeon crash with a different issue. I'm having the same exact issue and have tried all the same fixes. I can't find any instructions on how to check my version or to upgrade to a pre-release version. Can you please provide instructions for Ubuntu?

GiteaMirror commented

2026-04-12 10:59:25 -05:00

@dhiltgen commented on GitHub (Jan 24, 2024):

I'm having the same exact issue and have tried all the same fixes. I can't find any instructions on how to check my version or to upgrade to a pre-release version. Can you please provide instructions for Ubuntu?

To do a quick test:

wget https://github.com/ollama/ollama/releases/download/v0.1.21/ollama-linux-amd64
chmod a+x ollama-linux-amd64
sudo systemctl stop ollama
OLLAMA_DEBUG=1 ./ollama-linux-amd64 serve

@dhiltgen commented on GitHub (Jan 24, 2024): > I'm having the same exact issue and have tried all the same fixes. I can't find any instructions on how to check my version or to upgrade to a pre-release version. Can you please provide instructions for Ubuntu? To do a quick test: ``` wget https://github.com/ollama/ollama/releases/download/v0.1.21/ollama-linux-amd64 chmod a+x ollama-linux-amd64 sudo systemctl stop ollama OLLAMA_DEBUG=1 ./ollama-linux-amd64 serve ```

GiteaMirror commented

2026-04-12 10:59:25 -05:00

@venturaEffect commented on GitHub (Jan 25, 2024):

@silvergasp looks like you hit a GPU out-of-memory on 0.1.20. We've added some fixes to 0.1.21 to improve low memory GPUs, but the algorithm still isn't quite perfect.

@venturaEffect if you can try with 0.1.21 and share the server logs that will help us understand if this is a known issue we're working on, or something new.

I've done that already and shared to @jmorganca some days ago on Discord. He is aware of it. The problem is that even downgrading to a version that doesn't give this error the problem I'm facing is that it doesn't work for RAGs because of it's context window limitation issue. This has made me look for an alternative with LlamaIndex using their custom models. In any case I would love to use Ollama and Langchain but having this big limitation for RAGs it isn't very useful.

@venturaEffect commented on GitHub (Jan 25, 2024): > @silvergasp looks like you hit a GPU out-of-memory on 0.1.20. We've added some fixes to [0.1.21](https://github.com/ollama/ollama/releases/tag/v0.1.21) to improve low memory GPUs, but the algorithm still isn't quite perfect. > > @venturaEffect if you can try with [0.1.21](https://github.com/ollama/ollama/releases/tag/v0.1.21) and share the server logs that will help us understand if this is a known issue we're working on, or something new. I've done that already and shared to @jmorganca some days ago on Discord. He is aware of it. The problem is that even downgrading to a version that doesn't give this error the problem I'm facing is that it doesn't work for RAGs because of it's context window limitation issue. This has made me look for an alternative with LlamaIndex using their custom models. In any case I would love to use Ollama and Langchain but having this big limitation for RAGs it isn't very useful.

GiteaMirror commented

2026-04-12 10:59:26 -05:00

@jmorganca commented on GitHub (Jan 25, 2024):

@venturaEffect I'm so sorry you hit an error with large context windows. Will be fixing this soon, keep an eye on https://github.com/ollama/ollama/issues/1952

@jmorganca commented on GitHub (Jan 25, 2024): @venturaEffect I'm so sorry you hit an error with large context windows. Will be fixing this soon, keep an eye on https://github.com/ollama/ollama/issues/1952

GiteaMirror commented

2026-04-12 10:59:26 -05:00

@venturaEffect commented on GitHub (Jan 25, 2024):

@venturaEffect I'm so sorry you hit an error with large context windows. Will be fixing this soon, keep an eye on https://github.com/ollama/ollama/issues/1952

Following

@venturaEffect commented on GitHub (Jan 25, 2024): > @venturaEffect I'm so sorry you hit an error with large context windows. Will be fixing this soon, keep an eye on https://github.com/ollama/ollama/issues/1952 Following

GiteaMirror commented

2026-04-12 10:59:26 -05:00

@basillicus commented on GitHub (Jan 26, 2024):

Is there any solution that you know that could solve the issue?

It is not a solution but a workaround I am using until the bug is solved. I believe the problem is that ollama offloads more layers to the GPU than it will be able to handle. So I just trial-and-error change the number of layers to be offloaded to the GPU manually for the model you want to use until the model works.

ollama show dolphin-mistral --modelfile

will show the Modelfile of the model. I just use this Modelfile using FROM dolphin-mistral as the base model and adding PARAMETER num_gpu x

Then create the model:
ollama create dolphin-mistral_numGPU -f Modelfile_num_gpu_x
And keep modifying x until the model works.

EDIT: version 0.1.22 fixes my problem of offloading too many layers to the GPU.

@basillicus commented on GitHub (Jan 26, 2024): > Is there any solution that you know that could solve the issue? It is not a solution but a workaround I am using until the bug is solved. I believe the problem is that ollama offloads more layers to the GPU than it will be able to handle. So I just trial-and-error change the number of layers to be offloaded to the GPU manually for the model you want to use until the model works. `ollama show dolphin-mistral --modelfile` will show the Modelfile of the model. I just use this Modelfile using `FROM dolphin-mistral `as the base model and adding `PARAMETER num_gpu x` Then create the model: `ollama create dolphin-mistral_numGPU -f Modelfile_num_gpu_x ` And keep modifying x until the model works. EDIT: version 0.1.22 fixes my problem of offloading too many layers to the GPU.

GiteaMirror commented

2026-04-12 10:59:27 -05:00

@venturaEffect commented on GitHub (Jan 26, 2024):

Is there any solution that you know that could solve the issue?

It is not a solution but a workaround I am using until the bug is solved. I believe the problem is that ollama offloads more layers to the GPU than it will be able to handle. So I just trial-and-error change the number of layers to be offloaded to the GPU manually for the model you want to use until the model works.

ollama show dolphin-mistral --modelfile

will show the Modelfile of the model. I just use this Modelfile using FROM dolphin-mistral as the base model and adding PARAMETER num_gpu x

Then create the model:
ollama create dolphin-mistral_numGPU -f Modelfile_num_gpu_x
And keep modifying x until the model works.

Thanks, but this wouldn't solve the problem of context window limitation for RAGs with Ollama and Langchain I guess. It is just for the issue with the last Ollama version.

@venturaEffect commented on GitHub (Jan 26, 2024): > > > Is there any solution that you know that could solve the issue? > > It is not a solution but a workaround I am using until the bug is solved. I believe the problem is that ollama offloads more layers to the GPU than it will be able to handle. So I just trial-and-error change the number of layers to be offloaded to the GPU manually for the model you want to use until the model works. > > `ollama show dolphin-mistral --modelfile` > > will show the Modelfile of the model. I just use this Modelfile using `FROM dolphin-mistral `as the base model and adding `PARAMETER num_gpu x` > > Then create the model: > `ollama create dolphin-mistral_numGPU -f Modelfile_num_gpu_x > ` > And keep modifying x until the model works. > > Thanks, but this wouldn't solve the problem of context window limitation for RAGs with Ollama and Langchain I guess. It is just for the issue with the last Ollama version.

GiteaMirror commented

2026-04-12 10:59:27 -05:00

@nathaniel-brough commented on GitHub (Jan 26, 2024):

@silvergasp looks like you hit a GPU out-of-memory on 0.1.20. We've added some fixes to 0.1.21 to improve low memory GPUs, but the algorithm still isn't quite perfect.

Just tried the latest docker image. Looks like it's working fine for me now. Many thanks!

@nathaniel-brough commented on GitHub (Jan 26, 2024): > @silvergasp looks like you hit a GPU out-of-memory on 0.1.20. We've added some fixes to [0.1.21](https://github.com/ollama/ollama/releases/tag/v0.1.21) to improve low memory GPUs, but the algorithm still isn't quite perfect. Just tried the latest docker image. Looks like it's working fine for me now. Many thanks!

GiteaMirror referenced this issue

2026-04-22 02:38:10 -05:00

[GH-ISSUE #1214] Cache models for system restarts to not download again in docker #26380

GiteaMirror referenced this issue

2026-04-28 03:20:44 -05:00

[GH-ISSUE #1214] Cache models for system restarts to not download again in docker #47131

GiteaMirror referenced this issue

2026-05-03 09:54:36 -05:00

[GH-ISSUE #1214] Cache models for system restarts to not download again in docker #62656

Sign in to join this conversation.

Branches Tags

main

parth-mlx-decode-checkpoints

dhiltgen/ci

hoyyeva/editor-config-repair

parth-launch-codex-app

hoyyeva/fix-codex-model-metadata-warning

hoyyeva/qwen

hoyyeva/launch-backup-ux

parth/hide-claude-desktop-till-release

hoyyeva/opencode-image-modality

parth-add-claude-code-autoinstall

release_v0.22.0

pdevine/manifest-list

codex/fix-codex-model-metadata-warning

pdevine/addressable-manifest

brucemacd/launch-fetch-reccomended

jmorganca/llama-compat

launch-copilot-cli

hoyyeva/opencode-thinking

release_v0.20.7

parth-auto-save-backup

parth-test

jmorganca/gemma4-audio-replacements

fix-manifest-digest-on-pull

hoyyeva/vscode-improve

brucemacd/install-server-wait

brucemacd/download-before-remove

parth/update-claude-docs

parth-anthropic-reference-images-path

brucemac/start-ap-install

pdevine/mlx-update

pdevine/qwen35_vision

drifkin/api-show-fallback

mintlify/image-generation-1773352582

hoyyeva/server-context-length-local-config

jmorganca/faster-reptition-penalties

jmorganca/convert-nemotron

parth-pi-thinking

pdevine/sampling-penalties

jmorganca/fix-create-quantization-memory

dongchen/resumable_transfer_fix

pdevine/sampling-cache-error

jessegross/mlx-usage

hoyyeva/openclaw-config

hoyyeva/app-html

pdevine/qwen3next

brucemacd/sign-sh-install

brucemacd/tui-update

brucemacd/usage-api

jmorganca/launch-empty

fix-app-dist-embed

mxyng/mlx-compile

mxyng/mlx-quant

mxyng/mlx-glm4.7

mxyng/mlx

brucemacd/simplify-model-picker

jmorganca/qwen3-concurrent

fix-glm-4.7-flash-mla-config

drifkin/qwen3-coder-opening-tag

brucemacd/usage-cli

fix-cuda12-fattn-shmem

ollama-imagegen-docs

parth/fix-multiline-inputs

brucemacd/config-docs

mxyng/model-files

mxyng/simple-execute

fix-imagegen-ollama-models

mxyng/async-upload

jmorganca/lazy-no-dtype-changes

imagegen-auto-detect-create

parth/decrease-concurrent-download-hf

fix-mlx-quantize-init

jmorganca/x-cleanup

usage

imagegen-readme

jmorganca/glm-image

mlx-gpu-cd

jmorganca/imagegen-modelfile

parth/agent-skills

parth/agent-allowlist

parth/signed-in-offline

parth/agents

parth/fix-context-chopping

improve-cloud-flow

parth/add-models-websearch

parth/prompt-renderer-mcp

jmorganca/native-settings

jmorganca/download-stream-hash

jmorganca/client2-rebased

brucemacd/oai-chat-req-multipart

jessegross/multi_chunk_reserve

grace/additional-omit-empty

grace/mistral-3-large

mxyng/tokenizer2

mxyng/tokenizer

jessegross/flash

hoyyeva/windows-nacked-app

mxyng/cleanup-attention

grace/deepseek-parser

hoyyeva/remember-unsent-prompt

parth/add-lfs-pointer-error-conversion

parth/olmo2-test2

hoyyeva/ollama-launchagent-plist

nicole/olmo-model

parth/olmo-test

mxyng/remove-embedded

parth/render-template

jmorganca/intellect-3

parth/remove-prealloc-linter

jmorganca/cmd-eval

nicole/nomic-embed-text-fix

mxyng/lint-2

hoyyeva/add-gemini-3-pro-preview

hoyyeva/load-model-list

mxyng/expand-path

mxyng/environ-2

hoyyeva/deeplink-json-encoding

parth/improve-tool-calling-tests

hoyyeva/conversation

hoyyeva/assistant-edit-response

hoyyeva/thinking

origin/brucemacd/invalid-char-i-err

parth/improve-tool-calling

jmorganca/required-omitempty

grace/qwen3-vl-tests

mxyng/iter-client

parth/docs-readme

nicole/embed-test

pdevine/integration-benchstat

parth/remove-generate-cmd

parth/add-toolcall-id

mxyng/server-tests

jmorganca/glm-4.6

jmorganca/gin-h-compat

drifkin/stable-tool-args

pdevine/qwen3-more-thinking

parth/add-websearch-client

nicole/websearch_local

jmorganca/qwen3-coder-updates

grace/deepseek-v3-migration-tests

mxyng/fix-create

jmorganca/cloud-errors

pdevine/parser-tidy

revert-12233-parth/simplify-entrypoints-runner

parth/enable-so-gpt-oss

brucemacd/qwen3vl

jmorganca/readme-simplify

parth/gpt-oss-structured-outputs

revert-12039-jmorganca/tools-braces

mxyng/embeddings

mxyng/gguf

mxyng/benchmark

mxyng/types-null

parth/move-parsing

mxyng/gemma2

jmorganca/docs

mxyng/16-bit

mxyng/create-stdin

pdevine/authorizedkeys

mxyng/quant

parth/opt-in-error-context-window

brucemacd/cache-models

brucemacd/runner-completion

jmorganca/llama-update-6

brucemacd/benchmark-list

brucemacd/partial-read-caps

parth/deepseek-r1-tools

mxyng/omit-array

parth/tool-prefix-temp

brucemacd/runner-test

jmorganca/qwen25vl

brucemacd/model-forward-test-ext

parth/python-function-parsing

jmorganca/cuda-compression-none

drifkin/num-parallel

drifkin/chat-truncation-fix

jmorganca/sync

parth/python-tools-calling

drifkin/array-head-count

brucemacd/create-no-loop

parth/server-enable-content-stream-with-tools

qwen25omni

mxyng/v3

brucemacd/ropeconfig

jmorganca/silence-tokenizer

parth/sample-so-test

parth/sampling-structured-outputs

brucemacd/doc-go-engine

parth/constrained-sampling-json

jmorganca/mistral-wip

brucemacd/mistral-small-convert

parth/sample-unmarshal-json-for-params

brucemacd/jomorganca/mistral

pdevine/bfloat16

jmorganca/mistral

brucemacd/mistral

pdevine/logging

parth/sample-correctness-fix

parth/sample-fix-sorting

jmorgan/sample-fix-sorting-extras

jmorganca/temp-0-images

brucemacd/parallel-embed-models

brucemacd/shim-grammar

jmorganca/fix-gguf-error

bmizerany/nameswork

jmorganca/faster-releases

bmizerany/validatenames

brucemacd/err-no-vocab

brucemacd/rope-config

brucemacd/err-hint

brucemacd/qwen2_5

brucemacd/logprobs

brucemacd/new_runner_graph_bench

progress-flicker

brucemacd/forward-test

brucemacd/go_qwen2

pdevine/gemma2

jmorganca/add-missing-symlink-eval

mxyng/next-debug

parth/set-context-size-openai

brucemacd/next-bpe-bench

brucemacd/next-bpe-test

brucemacd/new_runner_e2e

brucemacd/new_runner_qwen2

pdevine/convert-cohere2

brucemacd/convert-cli

parth/log-probs

mxyng/next-mlx

mxyng/cmd-history

parth/templating

parth/tokenize-detokenize

brucemacd/check-key-register

bmizerany/grammar

jmorganca/vendor-081b29bd

mxyng/func-checks

jmorganca/fix-null-format

parth/fix-default-to-warn-json

jmorganca/qwen2vl

jmorganca/no-concat

parth/cmd-cleanup-SO

brucemacd/check-key-register-structured-err

parth/openai-stream-usage

parth/fix-referencing-so

stream-tools-stop

jmorganca/degin-1

brucemacd/install-path-clean

brucemacd/push-name-validation

brucemacd/browser-key-register

jmorganca/openai-fix-first-message

jmorganca/fix-proxy

jessegross/sample

parth/disallow-streaming-tools

dhiltgen/remove_submodule

jmorganca/ga

jmorganca/mllama

pdevine/newlines

pdevine/geems-2b

jmorganca/llama-bump

mxyng/modelname-7

mxyng/gin-slog

mxyng/modelname-6

jyan/convert-prog

jyan/quant5

paligemma-support

pdevine/import-docs

jmorganca/openai-context

jyan/paligemma

jyan/p2

jyan/palitest

bmizerany/embedspeedup

jmorganca/llama-vit

brucemacd/allow-ollama

royh/ep-methods

royh/whisper

mxyng/api-models

mxyng/fix-memory

jyan/q4_4/8

jyan/ollama-v

royh/stream-tools

roy-embed-parallel

bmizerany/hrm

revert-5963-revert-5924-mxyng/llama3.1-rope

royh/embed-viz

jyan/local2

jyan/auth

jyan/local

jyan/parse-temp

jmorganca/template-mistral

jyan/reord-g

royh-openai-suffixdocs

royh-imgembed

royh-embed-parallel

jyan/quant4

royh-precision

jyan/progress

pdevine/fix-template

jyan/quant3

pdevine/ggla

mxyng/update-registry-domain

jmorganca/ggml-static

mxyng/create-context

jyan/v0.146

mxyng/layers-from-files

build_dist

bmizerany/noseek

royh-ls

royh-name

timeout

mxyng/server-timestamp

bmizerany/nosillyggufslurps

royh-params

jmorganca/llama-cpp-7c26775

royh-openai-delete

royh-show-rigid

jmorganca/enable-fa

jmorganca/no-error-template

jyan/format

royh-testdelete

bmizerany/fastverify

language_support

pdevine/ps-glitches

brucemacd/tokenize

bruce/iq-quants

bmizerany/filepathwithcoloninhost

mxyng/split-bin

bmizerany/client-registry

jmorganca/if-none-match

native

jmorganca/native

jmorganca/batch-embeddings

jmorganca/initcmake

jmorganca/mm

pdevine/showggmlinfo

modenameenforcealphanum

bmizerany/modenameenforcealphanum

jmorganca/done-reason

jmorganca/llama-cpp-8960fe8

ollama.com

bmizerany/filepathnobuild

bmizerany/types/model/defaultfix

rmdisplaylong

nogogen

bmizerany/x

modelfile-readme

bmizerany/replacecolon

jmorganca/limit

jmorganca/execstack

jmorganca/replace-assets

mxyng/tune-concurrency

jmorganca/testing

whitespace-detection

jmorganca/options

upgrade-all

scratch

cuda-search

mattw/airenamer

mattw/allmodelsonhuggingface

mattw/quantcontext

mattw/whatneedstorun

brucemacd/llama-mem-calc

mattw/faq-context

mattw/communitylinks

mattw/noprune

mattw/python-functioncalling

rename

mxyng/install

pulse

remove-first

editor

mattw/selfqueryingretrieval

cgo

mattw/howtoquant

api

matt/streamingapi

format-config

mxyng/extra-args

shell

update-nous-hermes

cp-model

upload-progress

fix-unknown-model

fix-model-names

delete-fix

insecure-registry

ls

deletemodels

progressbar

readme-updates

license-layers

skip-list

list-models

modelpath

matt/examplemodelfiles

distribution

go-opts

1 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: github-starred/ollama#1214