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 #4100] Error: do encode request: Post "http://127.0.0.1:39207/tokenize": EOF #28307

New Issue
Open
opened 2026-04-22 06:20:20 -05:00 by GiteaMirror · 1 comment
GiteaMirror commented 2026-04-22 06:20:20 -05:00
Owner
Copy Link

Originally created by @j2l on GitHub (May 2, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/4100

What is the issue?

Hello,
I downloaded the Q4KM model from https://huggingface.co/LiteLLMs/French-Alpaca-Llama3-8B-Instruct-v1.0-GGUF/tree/main/Q4_K_M
renamed locally to French-Alpaca-Llama3-8B-Instruct-v1.gguf
Modelfile:
FROM "./French-Alpaca-Llama3-8B-Instruct-v1.gguf"
ollama create frll3 -f ./Modelfile

transferring model data 
creating model layer 
using already created layer sha256:08941f7a82566ca0116881e211330eae5838c20146132ed8fb9de46b6f5ea54b 
writing layer sha256:9f194159c3b80adee4448e1a1d380df743363881d417b5e9841e9611f884c155 
writing manifest 
success

ollama run frll3
>>> hi
Error: do encode request: Post "http://127.0.0.1:39207/tokenize": EOF

But ollama run llama3 works fine, I can discuss with it, no error. I have a 3060 (12GB VRAM)

Is it because of the model? The renaming? the Makefile?
Thanks

OS

Linux

GPU

Nvidia

CPU

Intel

Ollama version

0.1.32

Originally created by @j2l on GitHub (May 2, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/4100 ### What is the issue? Hello, I downloaded the Q4KM model from https://huggingface.co/LiteLLMs/French-Alpaca-Llama3-8B-Instruct-v1.0-GGUF/tree/main/Q4_K_M renamed locally to French-Alpaca-Llama3-8B-Instruct-v1.gguf Modelfile: `FROM "./French-Alpaca-Llama3-8B-Instruct-v1.gguf"` ollama create frll3 -f ./Modelfile ``` transferring model data creating model layer using already created layer sha256:08941f7a82566ca0116881e211330eae5838c20146132ed8fb9de46b6f5ea54b writing layer sha256:9f194159c3b80adee4448e1a1d380df743363881d417b5e9841e9611f884c155 writing manifest success ``` ollama run frll3 `>>> hi` `Error: do encode request: Post "http://127.0.0.1:39207/tokenize": EOF` But `ollama run llama3` works fine, I can discuss with it, no error. I have a 3060 (12GB VRAM) Is it because of the model? The renaming? the Makefile? Thanks ### OS Linux ### GPU Nvidia ### CPU Intel ### Ollama version 0.1.32
GiteaMirror added the bug label 2026-04-22 06:20:20 -05:00
GiteaMirror commented 2026-04-22 06:20:22 -05:00
Author
Owner
Copy Link

@Bonbon-Chan commented on GitHub (May 18, 2024):

Hi,

OLLama is working fine with all the "standard" model that I can access through ollama command line.

But I got the same problem with the same model :
`huggingface-cli download LiteLLMs/French-Alpaca-Llama3-8B-Instruct-v1.0-GGUF Q4_K_M/Q4_K_M-00001-of-00001.gguf --local-dir V:\ollama\downloads\French-Alpaca --local-dir-use-symlinks False

ollama create FrenchAlpaca -f modelfile

transferring model data
using existing layer sha256:08941f7a82566ca0116881e211330eae5838c20146132ed8fb9de46b6f5ea54b
creating new layer sha256:db8fbfd0cb288a053f83ac9014ca9bac2558b1bbcd80b5c408a548e7acba8a24
creating new layer sha256:1e7410908fd3e45ef2cd89cebf07322e0571fc4cda9e42ca7af366aa8d676f28
writing manifest
success

ollama run FrenchAlpaca

Salut !
Error: do encode request: Post "http://127.0.0.1:52278/tokenize": read tcp 127.0.0.1:52284->127.0.0.1:52278: wsarecv: Une connexion existante a dû être fermée par l’hôte distant.
`

With the modelfile:

FROM V:\ollama\downloads\French-Alpaca\Q4_K_M\Q4_K_M-00001-of-00001.gguf
PARAMETER num_ctx 32768

My configuration is a Windows 10 Intel with a Nvidia RTX3060.

I have try with the Q4_K_S model with the same result. I don't know if it is a problem when the model was created or a ollama problem but the huggingface page of the model claims that the model is compatible with ollama.

<!-- gh-comment-id:2118886382 --> @Bonbon-Chan commented on GitHub (May 18, 2024): Hi, OLLama is working fine with all the "standard" model that I can access through ollama command line. But I got the same problem with the same model : `huggingface-cli download LiteLLMs/French-Alpaca-Llama3-8B-Instruct-v1.0-GGUF Q4_K_M/Q4_K_M-00001-of-00001.gguf --local-dir V:\ollama\downloads\French-Alpaca --local-dir-use-symlinks False ollama create FrenchAlpaca -f modelfile transferring model data using existing layer sha256:08941f7a82566ca0116881e211330eae5838c20146132ed8fb9de46b6f5ea54b creating new layer sha256:db8fbfd0cb288a053f83ac9014ca9bac2558b1bbcd80b5c408a548e7acba8a24 creating new layer sha256:1e7410908fd3e45ef2cd89cebf07322e0571fc4cda9e42ca7af366aa8d676f28 writing manifest success ollama run FrenchAlpaca >>> Salut ! Error: do encode request: Post "http://127.0.0.1:52278/tokenize": read tcp 127.0.0.1:52284->127.0.0.1:52278: wsarecv: Une connexion existante a dû être fermée par l’hôte distant. ` With the modelfile: > FROM V:\ollama\downloads\French-Alpaca\Q4_K_M\Q4_K_M-00001-of-00001.gguf > PARAMETER num_ctx 32768 My configuration is a Windows 10 Intel with a Nvidia RTX3060. I have try with the Q4_K_S model with the same result. I don't know if it is a problem when the model was created or a ollama problem but the huggingface page of the model claims that the model is compatible with 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 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#28307
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?