github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 00:22:43 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #6265] Not a feature request, not a bug, problem with LLama3.1 #65959

New Issue
Closed
opened 2026-05-03 23:21:47 -05:00 by GiteaMirror · 8 comments
GiteaMirror commented 2026-05-03 23:21:47 -05:00
Owner
Copy Link

Originally created by @airdogvan on GitHub (Aug 8, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/6265

I have several models all running fine on ollama, including Llama3. Lama3.1 gives very long answers, then repeats them and finally prints random characters and if I didn't use the interface to stop seemingly would go on forever.

Running llama3.1 with the same parameters that seem to be ok with all other models.

Any suggestions welcome as it seems to be quite a bit more powerful than all other models.

Originally created by @airdogvan on GitHub (Aug 8, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/6265 I have several models all running fine on ollama, including Llama3. Lama3.1 gives very long answers, then repeats them and finally prints random characters and if I didn't use the interface to stop seemingly would go on forever. Running llama3.1 with the same parameters that seem to be ok with all other models. Any suggestions welcome as it seems to be quite a bit more powerful than all other models.
GiteaMirror added the feature request label 2026-05-03 23:21:47 -05:00
GiteaMirror commented 2026-05-03 23:22:00 -05:00
Author
Owner
Copy Link

@igorschlum commented on GitHub (Aug 8, 2024):

@airdogvan did you updated Ollama to version 0.3.4 and pull again your llama3.1 models? You should try and please share your ollama logs to help finding the problem if still present.

<!-- gh-comment-id:2276752808 --> @igorschlum commented on GitHub (Aug 8, 2024): @airdogvan did you updated Ollama to version 0.3.4 and pull again your llama3.1 models? You should try and please share your [ollama logs](https://github.com/ollama/ollama/blob/main/docs/troubleshooting.md) to help finding the problem if still present.
GiteaMirror commented 2026-05-03 23:22:01 -05:00
Author
Owner
Copy Link

@airdogvan commented on GitHub (Aug 8, 2024):

thanks for the quick answer. Running docker ollama. Had 3.2 pulled 3.4. Deleted llama3.1, restarted ollama. When I try to download llama3.1 error message:

The model you are attempting to pull requires a newer version of Ollama. Please download the latest version at: https://ollama.com/download

Double checked, have 3.4 running.

Not sure if uselful, but ollama logs:

ollama        | time=2024-08-08T22:14:23.129Z level=INFO source=images.go:781 msg="total blobs: 0"
ollama        | time=2024-08-08T22:14:23.131Z level=INFO source=images.go:788 msg="total unused blobs removed: 0"
ollama        | time=2024-08-08T22:14:23.131Z level=INFO source=routes.go:1155 msg="Listening on [::]:11434 (version 0.3.4)"
ollama        | time=2024-08-08T22:14:23.132Z level=INFO source=payload.go:30 msg="extracting embedded files" dir=/tmp/ollama143595698/runners
ollama        | time=2024-08-08T22:14:26.390Z level=INFO source=payload.go:44 msg="Dynamic LLM libraries [cpu cpu_avx cpu_avx2 cuda_v11 rocm_v60102]"
ollama        | time=2024-08-08T22:14:26.390Z level=INFO source=gpu.go:204 msg="looking for compatible GPUs"
ollama        | time=2024-08-08T22:14:26.487Z level=INFO source=types.go:105 msg="inference compute" id=GPU-828785ff-9ebf-6931-178f-d47c5df42ff3 library=cuda compute=8.6 driver=12.2 name="NVIDIA GeForce RTX 3060" total="11.8 GiB" available="4.8 GiB"
ollama        | [GIN] 2024/08/08 - 22:16:00 | 200 |      44.488��s |       127.0.0.1 | GET      "/api/version"
ollama        | 2024/08/08 22:17:10 routes.go:1108: INFO server config env="map[CUDA_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_VERSION: OLLAMA_DEBUG:false OLLAMA_FLASH_ATTENTION:false OLLAMA_HOST:http://0.0.0.0:11434 OLLAMA_INTEL_GPU:false OLLAMA_KEEP_ALIVE:5m0s OLLAMA_LLM_LIBRARY: OLLAMA_MAX_LOADED_MODELS:0 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:/root/.ollama/models 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://*] OLLAMA_RUNNERS_DIR: OLLAMA_SCHED_SPREAD:false OLLAMA_TMPDIR: ROCR_VISIBLE_DEVICES:]"
ollama        | time=2024-08-08T22:17:10.988Z level=INFO source=images.go:781 msg="total blobs: 0"
ollama        | time=2024-08-08T22:17:10.990Z level=INFO source=images.go:788 msg="total unused blobs removed: 0"
ollama        | time=2024-08-08T22:17:10.990Z level=INFO source=routes.go:1155 msg="Listening on [::]:11434 (version 0.3.4)"
ollama        | time=2024-08-08T22:17:10.991Z level=INFO source=payload.go:30 msg="extracting embedded files" dir=/tmp/ollama2816128774/runners
ollama        | time=2024-08-08T22:17:14.157Z level=INFO source=payload.go:44 msg="Dynamic LLM libraries [cpu cpu_avx cpu_avx2 cuda_v11 rocm_v60102]"
ollama        | time=2024-08-08T22:17:14.158Z level=INFO source=gpu.go:204 msg="looking for compatible GPUs"
ollama        | time=2024-08-08T22:17:14.245Z level=INFO source=types.go:105 msg="inference compute" id=GPU-828785ff-9ebf-6931-178f-d47c5df42ff3 library=cuda compute=8.6 driver=12.2 name="NVIDIA GeForce RTX 3060" total="11.8 GiB" available="4.4 GiB"
<!-- gh-comment-id:2276781901 --> @airdogvan commented on GitHub (Aug 8, 2024): thanks for the quick answer. Running docker ollama. Had 3.2 pulled 3.4. Deleted llama3.1, restarted ollama. When I try to download llama3.1 error message: ```The model you are attempting to pull requires a newer version of Ollama. Please download the latest version at: https://ollama.com/download ``` Double checked, have 3.4 running. Not sure if uselful, but ollama logs: ```ollama | 2024/08/08 22:14:23 routes.go:1108: INFO server config env="map[CUDA_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_VERSION: OLLAMA_DEBUG:false OLLAMA_FLASH_ATTENTION:false OLLAMA_HOST:http://0.0.0.0:11434 OLLAMA_INTEL_GPU:false OLLAMA_KEEP_ALIVE:5m0s OLLAMA_LLM_LIBRARY: OLLAMA_MAX_LOADED_MODELS:0 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:/root/.ollama/models 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://*] OLLAMA_RUNNERS_DIR: OLLAMA_SCHED_SPREAD:false OLLAMA_TMPDIR: ROCR_VISIBLE_DEVICES:]" ollama | time=2024-08-08T22:14:23.129Z level=INFO source=images.go:781 msg="total blobs: 0" ollama | time=2024-08-08T22:14:23.131Z level=INFO source=images.go:788 msg="total unused blobs removed: 0" ollama | time=2024-08-08T22:14:23.131Z level=INFO source=routes.go:1155 msg="Listening on [::]:11434 (version 0.3.4)" ollama | time=2024-08-08T22:14:23.132Z level=INFO source=payload.go:30 msg="extracting embedded files" dir=/tmp/ollama143595698/runners ollama | time=2024-08-08T22:14:26.390Z level=INFO source=payload.go:44 msg="Dynamic LLM libraries [cpu cpu_avx cpu_avx2 cuda_v11 rocm_v60102]" ollama | time=2024-08-08T22:14:26.390Z level=INFO source=gpu.go:204 msg="looking for compatible GPUs" ollama | time=2024-08-08T22:14:26.487Z level=INFO source=types.go:105 msg="inference compute" id=GPU-828785ff-9ebf-6931-178f-d47c5df42ff3 library=cuda compute=8.6 driver=12.2 name="NVIDIA GeForce RTX 3060" total="11.8 GiB" available="4.8 GiB" ollama | [GIN] 2024/08/08 - 22:16:00 | 200 | 44.488��s | 127.0.0.1 | GET "/api/version" ollama | 2024/08/08 22:17:10 routes.go:1108: INFO server config env="map[CUDA_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_VERSION: OLLAMA_DEBUG:false OLLAMA_FLASH_ATTENTION:false OLLAMA_HOST:http://0.0.0.0:11434 OLLAMA_INTEL_GPU:false OLLAMA_KEEP_ALIVE:5m0s OLLAMA_LLM_LIBRARY: OLLAMA_MAX_LOADED_MODELS:0 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:/root/.ollama/models 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://*] OLLAMA_RUNNERS_DIR: OLLAMA_SCHED_SPREAD:false OLLAMA_TMPDIR: ROCR_VISIBLE_DEVICES:]" ollama | time=2024-08-08T22:17:10.988Z level=INFO source=images.go:781 msg="total blobs: 0" ollama | time=2024-08-08T22:17:10.990Z level=INFO source=images.go:788 msg="total unused blobs removed: 0" ollama | time=2024-08-08T22:17:10.990Z level=INFO source=routes.go:1155 msg="Listening on [::]:11434 (version 0.3.4)" ollama | time=2024-08-08T22:17:10.991Z level=INFO source=payload.go:30 msg="extracting embedded files" dir=/tmp/ollama2816128774/runners ollama | time=2024-08-08T22:17:14.157Z level=INFO source=payload.go:44 msg="Dynamic LLM libraries [cpu cpu_avx cpu_avx2 cuda_v11 rocm_v60102]" ollama | time=2024-08-08T22:17:14.158Z level=INFO source=gpu.go:204 msg="looking for compatible GPUs" ollama | time=2024-08-08T22:17:14.245Z level=INFO source=types.go:105 msg="inference compute" id=GPU-828785ff-9ebf-6931-178f-d47c5df42ff3 library=cuda compute=8.6 driver=12.2 name="NVIDIA GeForce RTX 3060" total="11.8 GiB" available="4.4 GiB" ```
GiteaMirror commented 2026-05-03 23:22:02 -05:00
Author
Owner
Copy Link

@igorschlum commented on GitHub (Aug 9, 2024):

@airdogvan when you type

ollama --version

what is displayed? It is strange that ollama3.1 don't pull.

<!-- gh-comment-id:2277704459 --> @igorschlum commented on GitHub (Aug 9, 2024): @airdogvan when you type ```bash ollama --version ``` what is displayed? It is strange that ollama3.1 don't pull.
GiteaMirror commented 2026-05-03 23:22:03 -05:00
Author
Owner
Copy Link

@airdogvan commented on GitHub (Aug 9, 2024):

ollama --version
ollama version is 0.3.4

This seems to be a open_webui problem. I went in the container, ollama run llama3.1:latest, it downloaded it and, after loading, I could send queries that were answered correctly and no hallucinations or any other odd behavior.

<!-- gh-comment-id:2278327145 --> @airdogvan commented on GitHub (Aug 9, 2024): ollama --version ollama version is 0.3.4 This seems to be a open_webui problem. I went in the container, ollama run llama3.1:latest, it downloaded it and, after loading, I could send queries that were answered correctly and no hallucinations or any other odd behavior.
GiteaMirror commented 2026-05-03 23:22:04 -05:00
Author
Owner
Copy Link

@igorschlum commented on GitHub (Aug 9, 2024):

@airdogvan if it is an open_webui problem, please close this issue.
BTW new version of llama3.1 is being uploaded today in the library. Try also to pull again the llama3.1 model you are using.

<!-- gh-comment-id:2278785738 --> @igorschlum commented on GitHub (Aug 9, 2024): @airdogvan if it is an open_webui problem, please close this issue. BTW new version of llama3.1 is being uploaded today in the library. Try also to pull again the llama3.1 model you are using.
GiteaMirror commented 2026-05-03 23:22:05 -05:00
Author
Owner
Copy Link

@KrallBigFan commented on GitHub (Aug 11, 2024):

got the same issue on MacOS M3, (ollama version is 0.3.4, llama3:70b run ok):

~ ollama run llama3.1:70b pulling manifest Error: pull model manifest: 412: The model you are attempting to pull requires a newer version of Ollama. Please download the latest version at: https://ollama.com/download

in fact, this is what https://registry.ollama.ai/v2/library/llama3.1/manifests/70b returns.

<!-- gh-comment-id:2282632168 --> @KrallBigFan commented on GitHub (Aug 11, 2024): got the same issue on MacOS M3, (ollama version is 0.3.4, llama3:70b run ok): `~ ollama run llama3.1:70b pulling manifest Error: pull model manifest: 412: The model you are attempting to pull requires a newer version of Ollama. Please download the latest version at: https://ollama.com/download` in fact, this is what https://registry.ollama.ai/v2/library/llama3.1/manifests/70b returns.
GiteaMirror commented 2026-05-03 23:22:07 -05:00
Author
Owner
Copy Link

@igorschlum commented on GitHub (Aug 11, 2024):

@KrallBigFan can you check with

ollama --version

what is your version.
You should have

ollama version is 0.3.4

if ollama cannot pull your model, can you try

ollama rm llama3.1:70b

and then again

ollama pull llama3.1:70b

let us know if it works.

<!-- gh-comment-id:2282771360 --> @igorschlum commented on GitHub (Aug 11, 2024): @KrallBigFan can you check with ```bash ollama --version ``` what is your version. You should have ollama version is 0.3.4 if ollama cannot pull your model, can you try ```bash ollama rm llama3.1:70b ``` and then again ```bash ollama pull llama3.1:70b ``` let us know if it works.
GiteaMirror commented 2026-05-03 23:22:09 -05:00
Author
Owner
Copy Link

@KrallBigFan commented on GitHub (Aug 11, 2024):

no more issue after the reinstalling the brew package:
brew reinstall ollama
and
brew services restart ollama

<!-- gh-comment-id:2282876881 --> @KrallBigFan commented on GitHub (Aug 11, 2024): no more issue after the reinstalling the brew package: `brew reinstall ollama` and `brew services restart ollama`
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 feature request
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#65959
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?