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 #422] Error: Post "http://localhost:11434/api/generate": EOF with long propmts with phind-codellama #62230

New Issue
Closed
opened 2026-05-03 07:57:27 -05:00 by GiteaMirror · 20 comments
GiteaMirror commented 2026-05-03 07:57:27 -05:00
Owner
Copy Link

Originally created by @tomduncalf on GitHub (Aug 26, 2023).
Original GitHub issue: https://github.com/ollama/ollama/issues/422

It seems like if you provide a long prompt (I was using one of 1,000ish tokens according to OpenAI tokenizer) with this model, you get an error Error: Post "http://localhost:11434/api/generate": EOF. It may or may not relate to the contents of the prompt as well as the length

Originally created by @tomduncalf on GitHub (Aug 26, 2023). Original GitHub issue: https://github.com/ollama/ollama/issues/422 It seems like if you provide a long prompt (I was using one of 1,000ish tokens according to OpenAI tokenizer) with this model, you get an error `Error: Post "http://localhost:11434/api/generate": EOF`. It may or may not relate to the contents of the prompt as well as the length
GiteaMirror added the bug label 2026-05-03 07:57:27 -05:00
GiteaMirror commented 2026-05-03 07:57:32 -05:00
Author
Owner
Copy Link

@hmottestad commented on GitHub (Aug 26, 2023):

Here is an example:

ollama run phind-codellama " Finished the method marked as TODO in the following Java code:\n\n\npackage org.eclipse.rdf4j.sail.shacl.ast.constraintcomponents;\n\nimport java.util.Collections;\nimport java.util.HashSet;\nimport java.util.List;\nimport java.util.Set;\n\nimport org.eclipse.rdf4j.model.IRI;\nimport org.eclipse.rdf4j.model.Literal;\nimport org.eclipse.rdf4j.model.Model;\nimport org.eclipse.rdf4j.model.Resource;\nimport org.eclipse.rdf4j.model.Value;\nimport org.eclipse.rdf4j.model.vocabulary.SHACL;\nimport org.eclipse.rdf4j.sail.shacl.SourceConstraintComponent;\nimport org.eclipse.rdf4j.sail.shacl.ValidationSettings;\nimport org.eclipse.rdf4j.sail.shacl.ast.SparqlFragment;\nimport org.eclipse.rdf4j.sail.shacl.ast.StatementMatcher;\nimport org.eclipse.rdf4j.sail.shacl.ast.StatementMatcher.Variable;\nimport org.eclipse.rdf4j.sail.shacl.ast.ValidationApproach;\nimport org.eclipse.rdf4j.sail.shacl.ast.ValidationQuery;\nimport org.eclipse.rdf4j.sail.shacl.ast.paths.Path;\nimport org.eclipse.rdf4j.sail.shacl.ast.planNodes.BulkedExternalLeftOuterJoin;\nimport org.eclipse.rdf4j.sail.shacl.ast.planNodes.EmptyNode;\nimport org.eclipse.rdf4j.sail.shacl.ast.planNodes.GroupByFilter;\nimport org.eclipse.rdf4j.sail.shacl.ast.planNodes.PlanNode;\nimport org.eclipse.rdf4j.sail.shacl.ast.planNodes.PlanNodeProvider;\nimport org.eclipse.rdf4j.sail.shacl.ast.planNodes.ShiftToPropertyShape;\nimport org.eclipse.rdf4j.sail.shacl.ast.planNodes.TrimToTarget;\nimport org.eclipse.rdf4j.sail.shacl.ast.planNodes.UnBufferedPlanNode;\nimport org.eclipse.rdf4j.sail.shacl.ast.planNodes.UnionNode;\nimport org.eclipse.rdf4j.sail.shacl.ast.planNodes.Unique;\nimport org.eclipse.rdf4j.sail.shacl.ast.planNodes.ValidationTuple;\nimport org.eclipse.rdf4j.sail.shacl.ast.planNodes.ValueInFilter;\nimport org.eclipse.rdf4j.sail.shacl.ast.targets.EffectiveTarget;\nimport org.eclipse.rdf4j.sail.shacl.wrapper.data.ConnectionsGroup;\nimport org.eclipse.rdf4j.sail.shacl.wrapper.data.RdfsSubClassOfReasoner;\n\npublic class HasValueConstraintComponent extends AbstractConstraintComponent {\n\n	private final Value hasValue;\n\n	public HasValueConstraintComponent(Value hasValue) {\n		this.hasValue = hasValue;\n	}\n\n	@Override\n	public void toModel(Resource subject, IRI predicate, Model model, Set<Resource> cycleDetection) {\n		model.add(subject, SHACL.HAS_VALUE, hasValue);\n	}\n\n	@Override\n	public SourceConstraintComponent getConstraintComponent() {\n		return SourceConstraintComponent.HasValueConstraintComponent;\n	}\n\n	@Override\n	public ConstraintComponent deepClone() {\n		return new HasValueConstraintComponent(hasValue);\n	}\n\n	@Override\n	public PlanNode generateTransactionalValidationPlan(ConnectionsGroup connectionsGroup,\n			ValidationSettings validationSettings, PlanNodeProvider overrideTargetNode, Scope scope) {\n\n		StatementMatcher.StableRandomVariableProvider stableRandomVariableProvider = "
<!-- gh-comment-id:1694412020 --> @hmottestad commented on GitHub (Aug 26, 2023): Here is an example: ``` ollama run phind-codellama " Finished the method marked as TODO in the following Java code:\n\n\npackage org.eclipse.rdf4j.sail.shacl.ast.constraintcomponents;\n\nimport java.util.Collections;\nimport java.util.HashSet;\nimport java.util.List;\nimport java.util.Set;\n\nimport org.eclipse.rdf4j.model.IRI;\nimport org.eclipse.rdf4j.model.Literal;\nimport org.eclipse.rdf4j.model.Model;\nimport org.eclipse.rdf4j.model.Resource;\nimport org.eclipse.rdf4j.model.Value;\nimport org.eclipse.rdf4j.model.vocabulary.SHACL;\nimport org.eclipse.rdf4j.sail.shacl.SourceConstraintComponent;\nimport org.eclipse.rdf4j.sail.shacl.ValidationSettings;\nimport org.eclipse.rdf4j.sail.shacl.ast.SparqlFragment;\nimport org.eclipse.rdf4j.sail.shacl.ast.StatementMatcher;\nimport org.eclipse.rdf4j.sail.shacl.ast.StatementMatcher.Variable;\nimport org.eclipse.rdf4j.sail.shacl.ast.ValidationApproach;\nimport org.eclipse.rdf4j.sail.shacl.ast.ValidationQuery;\nimport org.eclipse.rdf4j.sail.shacl.ast.paths.Path;\nimport org.eclipse.rdf4j.sail.shacl.ast.planNodes.BulkedExternalLeftOuterJoin;\nimport org.eclipse.rdf4j.sail.shacl.ast.planNodes.EmptyNode;\nimport org.eclipse.rdf4j.sail.shacl.ast.planNodes.GroupByFilter;\nimport org.eclipse.rdf4j.sail.shacl.ast.planNodes.PlanNode;\nimport org.eclipse.rdf4j.sail.shacl.ast.planNodes.PlanNodeProvider;\nimport org.eclipse.rdf4j.sail.shacl.ast.planNodes.ShiftToPropertyShape;\nimport org.eclipse.rdf4j.sail.shacl.ast.planNodes.TrimToTarget;\nimport org.eclipse.rdf4j.sail.shacl.ast.planNodes.UnBufferedPlanNode;\nimport org.eclipse.rdf4j.sail.shacl.ast.planNodes.UnionNode;\nimport org.eclipse.rdf4j.sail.shacl.ast.planNodes.Unique;\nimport org.eclipse.rdf4j.sail.shacl.ast.planNodes.ValidationTuple;\nimport org.eclipse.rdf4j.sail.shacl.ast.planNodes.ValueInFilter;\nimport org.eclipse.rdf4j.sail.shacl.ast.targets.EffectiveTarget;\nimport org.eclipse.rdf4j.sail.shacl.wrapper.data.ConnectionsGroup;\nimport org.eclipse.rdf4j.sail.shacl.wrapper.data.RdfsSubClassOfReasoner;\n\npublic class HasValueConstraintComponent extends AbstractConstraintComponent {\n\n private final Value hasValue;\n\n public HasValueConstraintComponent(Value hasValue) {\n this.hasValue = hasValue;\n }\n\n @Override\n public void toModel(Resource subject, IRI predicate, Model model, Set<Resource> cycleDetection) {\n model.add(subject, SHACL.HAS_VALUE, hasValue);\n }\n\n @Override\n public SourceConstraintComponent getConstraintComponent() {\n return SourceConstraintComponent.HasValueConstraintComponent;\n }\n\n @Override\n public ConstraintComponent deepClone() {\n return new HasValueConstraintComponent(hasValue);\n }\n\n @Override\n public PlanNode generateTransactionalValidationPlan(ConnectionsGroup connectionsGroup,\n ValidationSettings validationSettings, PlanNodeProvider overrideTargetNode, Scope scope) {\n\n StatementMatcher.StableRandomVariableProvider stableRandomVariableProvider = " ```
GiteaMirror commented 2026-05-03 07:57:33 -05:00
Author
Owner
Copy Link

@matiasdominguez commented on GitHub (Aug 29, 2023):

I'm getting the same error without a long prompt, even a short hey will trigger the same Error: Post "http://localhost:11434/api/generate": EOF for me. Specs: 2021 M1 Pro / 32 GB RAM.

<!-- gh-comment-id:1696789018 --> @matiasdominguez commented on GitHub (Aug 29, 2023): I'm getting the same error without a long prompt, even a short `hey` will trigger the same `Error: Post "http://localhost:11434/api/generate": EOF` for me. Specs: 2021 M1 Pro / 32 GB RAM.
GiteaMirror commented 2026-05-03 07:57:34 -05:00
Author
Owner
Copy Link

@hmottestad commented on GitHub (Aug 29, 2023):

I'm getting the same error without a long prompt, even a short hey will trigger the same Error: Post "http://localhost:11434/api/generate": EOF for me. Specs: 2021 M1 Pro / 32 GB RAM.

If you installed Ollama through homebrew then you might have an old version. Have you tried with downloading the newest version directly from https://ollama.ai ?

<!-- gh-comment-id:1696882604 --> @hmottestad commented on GitHub (Aug 29, 2023): > I'm getting the same error without a long prompt, even a short `hey` will trigger the same `Error: Post "http://localhost:11434/api/generate": EOF` for me. Specs: 2021 M1 Pro / 32 GB RAM. If you installed Ollama through homebrew then you might have an old version. Have you tried with downloading the newest version directly from https://ollama.ai ?
GiteaMirror commented 2026-05-03 07:57:35 -05:00
Author
Owner
Copy Link

@matiasdominguez commented on GitHub (Aug 31, 2023):

Thanks @hmottestad. That actually fixed it, I thought the homebrew version was the latest one.

<!-- gh-comment-id:1700187945 --> @matiasdominguez commented on GitHub (Aug 31, 2023): Thanks @hmottestad. That actually fixed it, I thought the homebrew version was the latest one.
GiteaMirror commented 2026-05-03 07:57:35 -05:00
Author
Owner
Copy Link

@technovangelist commented on GitHub (Sep 1, 2023):

Sorry, we didn't create the homebrew package. It looks like its up to date, but maybe it doesn't automatically update like the one from the website does.

<!-- gh-comment-id:1702975342 --> @technovangelist commented on GitHub (Sep 1, 2023): Sorry, we didn't create the homebrew package. It looks like its up to date, but maybe it doesn't automatically update like the one from the website does.
GiteaMirror commented 2026-05-03 07:57:36 -05:00
Author
Owner
Copy Link

@hmottestad commented on GitHub (Sep 7, 2023):

This seems to be fixed in the latests v0.0.18 :)

<!-- gh-comment-id:1709716682 --> @hmottestad commented on GitHub (Sep 7, 2023): This seems to be fixed in the latests v0.0.18 :)
GiteaMirror commented 2026-05-03 07:57:36 -05:00
Author
Owner
Copy Link

@jmorganca commented on GitHub (Sep 7, 2023):

Fantastic. Closing this for now, but feel free to re-open if we see it again.

<!-- gh-comment-id:1709958371 --> @jmorganca commented on GitHub (Sep 7, 2023): Fantastic. Closing this for now, but feel free to re-open if we see it again.
GiteaMirror commented 2026-05-03 07:57:37 -05:00
Author
Owner
Copy Link

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

image
installer seems to give the same error

<!-- gh-comment-id:1883559616 --> @heapframe commented on GitHub (Jan 9, 2024): ![image](https://github.com/jmorganca/ollama/assets/63340001/f372eced-ac56-436b-adf6-048f45723fb2) installer seems to give the same error
GiteaMirror commented 2026-05-03 07:57:38 -05:00
Author
Owner
Copy Link

@gokulkumar3198 commented on GitHub (Jan 11, 2024):

I am also facing the same EOF error, could not figure out the problem. I have tried with the latest docker image as of today and v0.1.18 also.
Error: Post "http://0.0.0.0:11434/api/generate": EOF

<!-- gh-comment-id:1886985448 --> @gokulkumar3198 commented on GitHub (Jan 11, 2024): I am also facing the same EOF error, could not figure out the problem. I have tried with the latest docker image as of today and v0.1.18 also. **Error: Post "http://0.0.0.0:11434/api/generate": EOF**
GiteaMirror commented 2026-05-03 07:57:39 -05:00
Author
Owner
Copy Link

@nps798 commented on GitHub (Jan 16, 2024):

same problem here last week. Running ollama predefined model worked fine, but I faced issues when executing custom model (convert from makefile via -f command)

fortunately, I resolved it by relocating the origin makefile and GGUF file. Initially, the path containing these files was synced with my NAS which seemed to contribute to this EOF issue. Although I don't know the exact reason behind it, changing to another folder/path fixed the problem for me.

<!-- gh-comment-id:1893672990 --> @nps798 commented on GitHub (Jan 16, 2024): same problem here last week. Running ollama predefined model worked fine, but I faced issues when executing custom model (convert from makefile via -f command) fortunately, I resolved it by relocating the origin makefile and GGUF file. Initially, the path containing these files was synced with my NAS which seemed to contribute to this EOF issue. Although I don't know the exact reason behind it, changing to another folder/path fixed the problem for me.
GiteaMirror commented 2026-05-03 07:57:39 -05:00
Author
Owner
Copy Link

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

Same problem here, giving it a prompt longer than two words immediately results in: Error: Post "http://127.0.0.1:11434/api/generate": EOF

<!-- gh-comment-id:1910705283 --> @fabianslife commented on GitHub (Jan 25, 2024): Same problem here, giving it a prompt longer than two words immediately results in: Error: Post "http://127.0.0.1:11434/api/generate": EOF
GiteaMirror commented 2026-05-03 07:57:39 -05:00
Author
Owner
Copy Link

@alexhegit commented on GitHub (Feb 5, 2024):

I go a problem like this one with log :

Error: Post "http://localhost:11434/api/chat": EOF

<!-- gh-comment-id:1926456172 --> @alexhegit commented on GitHub (Feb 5, 2024): I go a problem like this one with log : Error: Post "http://localhost:11434/api/chat": EOF
GiteaMirror commented 2026-05-03 07:57:40 -05:00
Author
Owner
Copy Link

@mehdiataei commented on GitHub (Feb 21, 2024):

Still happening

<!-- gh-comment-id:1957659006 --> @mehdiataei commented on GitHub (Feb 21, 2024): Still happening
GiteaMirror commented 2026-05-03 07:57:41 -05:00
Author
Owner
Copy Link

@AntDX316 commented on GitHub (Feb 21, 2024):

I have the issue on gemma:7b but mistral one works?

<!-- gh-comment-id:1958230081 --> @AntDX316 commented on GitHub (Feb 21, 2024): I have the issue on gemma:7b but mistral one works?
GiteaMirror commented 2026-05-03 07:57:41 -05:00
Author
Owner
Copy Link

@jafarzzz commented on GitHub (Feb 22, 2024):

Facing the same issue with gemma:7b. Other models are working fine. I updated the ollama. Still the problem persists.

<!-- gh-comment-id:1958868115 --> @jafarzzz commented on GitHub (Feb 22, 2024): Facing the same issue with gemma:7b. Other models are working fine. I updated the ollama. Still the problem persists.
GiteaMirror commented 2026-05-03 07:57:42 -05:00
Author
Owner
Copy Link

@serhanylmz commented on GitHub (Feb 22, 2024):

have the same issue with gemma:7b, can confirm

<!-- gh-comment-id:1958897542 --> @serhanylmz commented on GitHub (Feb 22, 2024): have the same issue with gemma:7b, can confirm
GiteaMirror commented 2026-05-03 07:57:42 -05:00
Author
Owner
Copy Link

@benjamin-bertram commented on GitHub (Feb 22, 2024):

Same here

<!-- gh-comment-id:1959087720 --> @benjamin-bertram commented on GitHub (Feb 22, 2024): Same here
GiteaMirror commented 2026-05-03 07:57:43 -05:00
Author
Owner
Copy Link

@jafarzzz commented on GitHub (Feb 22, 2024):

I fixed it by upgrading the ollama to 0.1.26. You wont be able to do it from the application. Uninstall the ollama and download the latest one from: https://ollama.com/

gemma:7b worked after this fix.

<!-- gh-comment-id:1959185554 --> @jafarzzz commented on GitHub (Feb 22, 2024): I fixed it by upgrading the ollama to 0.1.26. You wont be able to do it from the application. Uninstall the ollama and download the latest one from: https://ollama.com/ gemma:7b worked after this fix.
GiteaMirror commented 2026-05-03 07:57:44 -05:00
Author
Owner
Copy Link

@fithisux commented on GitHub (Feb 28, 2024):

In Windows 10PRO after calling

resp = gemma_2b.complete("Who is Paul Graham?")
print(resp)

I have the same error. In logs

[GIN] 2024/02/28 - 13:21:17 | 404 |         709µs |       127.0.0.1 | POST     "/api/generate"

System:

Edition Windows 10 Pro
Version 22H2
Installed on ‎19/‎1/‎2024
OS build 19045.4046
Experience Windows Feature Experience Pack 1000.19053.1000.0

<!-- gh-comment-id:1968776941 --> @fithisux commented on GitHub (Feb 28, 2024): In Windows 10PRO after calling ``` resp = gemma_2b.complete("Who is Paul Graham?") print(resp) ``` I have the same error. In logs ``` [GIN] 2024/02/28 - 13:21:17 | 404 | 709µs | 127.0.0.1 | POST "/api/generate" ``` System: Edition Windows 10 Pro Version 22H2 Installed on ‎19/‎1/‎2024 OS build 19045.4046 Experience Windows Feature Experience Pack 1000.19053.1000.0
GiteaMirror commented 2026-05-03 07:57:44 -05:00
Author
Owner
Copy Link

@BestMat commented on GitHub (Apr 22, 2024):

Same here, any solution for this problem?

<!-- gh-comment-id:2068888912 --> @BestMat commented on GitHub (Apr 22, 2024): Same here, any solution for this problem?
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
amd
api
app
bug
build
cli
cloud
compatibility
context-length
create
docker
documentation
embeddings
feature request
feedback wanted
good first issue
gpt-oss
gpu
harmony
help wanted
image
install
intel
js
launch
linux
macos
memory
mlx
model
needs more info
networking
nvidia
ollama.com
performance
pull-request
Mirrored from GitHub Pull Request
 python
question
registry
rendering
thinking
tools
top
vulkan
windows
wsl
No Label bug
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
GiteaMirror ninjasurge
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#62230
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?