AI/Ollama: Difference between revisions

From Chorke Wiki
Jump to navigation Jump to search
← Older editNewer edit →
VisualWikitext
Line 190: Line 190:
setsid open http://127.0.0.1:11434 >/dev/null 2>&1 &
setsid open http://127.0.0.1:11434 >/dev/null 2>&1 &
</syntaxhighlight>
</syntaxhighlight>
 
|}
{|class='wikitable mw-collapsible mw-collapsed'
!scope='col' style='text-align:left'| Ollama » ArcGPU
|-
|valign='top'|
<syntaxhighlight lang='bash'>
<syntaxhighlight lang='bash'>
sudo journalctl -u ollama --no-pager | grep -i "sycl\|intel\|gpu"
sudo journalctl -u ollama --no-pager | grep -i "sycl\|intel\|gpu"
Line 227: Line 231:
sudo intel_gpu_top
sudo intel_gpu_top
clinfo | grep 'Device Name'
clinfo | grep 'Device Name'
</syntaxhighlight>
<syntaxhighlight lang='bash'>
cat <<'EXE' | sudo bash
rm /etc/apt/sources.list.d/intel-graphics.list
rm /etc/apt/keyrings/intel-graphics.asc
EXE
</syntaxhighlight>
</syntaxhighlight>
|}
|}

Revision as of 03:41, 8 March 2026

curl -fsSL https://ollama.com/install.sh | sh
ollama pull gpt-oss:20b
ollama --version
ollama ls

curl -fsSL https://claude.ai/install.sh  | bash
ollama launch claude --model gpt-oss:20b

export ANTHROPIC_BASE_URL=http://localhost:11434
export ANTHROPIC_AUTH_TOKEN=ollama
export ANTHROPIC_API_KEY=''

export OLLAMA_NUM_CTX=32768
export OLLAMA_KEEP_ALIVE=5m
claude --model gpt-oss:20b

Diagram 

@startuml
autonumber
skinparam backgroundColor    transparent
skinparam DefaultFontName    Helvetica
skinparam actorStyle         awesome
skinparam ParticipantPadding 20
skinparam BoxPadding         10

title Claude ↔ Local FS ↔ Ollama ↔ GPT-OSS:20b

actor "Developer"                  as dev

box "Local Development PC" #LightBlue
    participant "Claude Code CLI"  as claude
    participant "Local Filesystem" as fs
end box

box "Kubernetes Cluster (K3s)" #Yellow
    participant "Ollama Service"   as ollama
    participant "GPT-OSS:20b"      as model
end box

dev      -> claude : Runs "claude --model gpt-oss:20b"
claude   -> fs     : Scans repository context
fs      --> claude : File contents / Git history

claude   -> ollama : POST /v1/messages (Anthropic API)
note right: Payload includes system prompt \nand local code context

ollama   -> model  : Load weights into GPU VRAM
model   --> ollama : Inference processing...

ollama -->> claude : Streamed Response (Tokens)
claude   -> dev    : Displays suggested code changes

@enduml

Sturcture 

@startsalt
skinparam backgroundColor transparent
skinparam defaultFontName monospaced
{
{T-
+**/**                       | Root File System
++**/usr/local/bin/**        | Executable Binaries
+++ollama                    | Ollama Server (Standalone)
++**/etc/systemd/system/**   | Systemd Services
+++ollama.service            | Systemd service file
++**/home/$USER/**           | User's Home Directory
+++**.local/bin/**           | User's Executable Binaries
++++claude                   | Claude Code CLI
+++**.ollama/**              | Ollama Data Directory
++++history                  | CLI Chat History
++++**models/**              | Saved Models
+++++blobs/                  | Weights **(gpt-oss:20b)**
+++++manifests/              | Model metadata
+++**.claude/**              | Claude Code Data Directory
++++config.json              | API URL, keys, project context
++++memory/                  | Persistent memory
+++**my-project/**           | Your development folder
++++.claude/                 | Project specific settings
++++CLAUDE.md                | Guidebook for current project
}
}
@endsalt

Optimization

Optimization

Variable Value Impact
OLLAMA_FLASH_ATTENTION 1 Reduces memory usage and speeds up processing for long code files. Highly recommended for coding.
OLLAMA_KV_CACHE_TYPE q8_0 or q4_0 Compresses the short-term memory cache. q8_0 saves space with almost no quality loss; q4_0 saves even more space.
OLLAMA_NUM_PARALLEL 1 Crucial for 32GB RAM. Limits Ollama to one task at a time to prevent Out of Memory crashes when using a 20B model.
OLLAMA_KEEP_ALIVE 30m Keeps the 20B model in your RAM for 30 minutes after use so you don't have to wait 20 seconds for it to reload every time.
OLLAMA_NUM_CTX 6384 to 32768 The most important. Controls the brain capacity. 32k is standard for Claude Code but uses ~3GB more RAM than the default 4k.
OLLAMA_NUM_GPU 999 Forces Ollama to offload as many layers as possible to your Intel Arc iGPU instead of the slower CPU.
OLLAMA_NUM_GPU 0 Forces Ollama CPU-only for all models
CUDA_VISIBLE_DEVICES -1 Hides any GPU from the Ollama process
ROCR_VISIBLE_DEVICES -1 Hides any GPU from the Ollama process
OLLAMA_INTEL_GPU 1 Forces Ollama to use Intel Arc GPU

Installation

Installation

Yoga Pro 7i (G9 + U7 155H + 32GB + 1TB)

Ollama » Install 
curl -fsSL https://ollama.com/install.sh | sh
:'
>>> Installing ollama to /usr/local
[sudo] password for shahed: 
>>> Downloading ollama-linux-amd64.tar.zst
######################################################################## 100.0%
>>> Creating ollama user...
>>> Adding ollama user to render group...
>>> Adding ollama user to video group...
>>> Adding current user to ollama group...
>>> Creating ollama systemd service...
>>> Enabling and starting ollama service...
Created symlink /etc/systemd/system/default.target.wants/ollama.service → /etc/systemd/system/ollama.service.
>>> The Ollama API is now available at 127.0.0.1:11434.
>>> Install complete. Run "ollama" from the command line.
WARNING: No NVIDIA/AMD GPU detected. Ollama will run in CPU-only mode.
'

sudo systemctl daemon-reload 
sudo systemctl restart  ollama.service 
sudo systemctl status   ollama.service

cat /etc/systemd/system/ollama.service
setsid open http://127.0.0.1:11434 >/dev/null 2>&1 &
Ollama » ArcGPU 
sudo journalctl -u ollama --no-pager | grep -i "sycl\|intel\|gpu"
sudo apt update && sudo apt install intel-gpu-tools
sudo intel_gpu_top

cat <<'EXE' | sudo bash
apt-get update;echo
apt-get install -y apt-transport-https ca-certificates gnupg build-essential
apt-get install -y software-properties-common git curl file procps libfuse2
apt-get clean;sleep 5
EXE

curl -fsSL https://repositories.intel.com/gpu/intel-graphics.key \
 | sudo tee /etc/apt/keyrings/intel-graphics.asc >/dev/null

DISTRIBUTION=$(. /etc/os-release && echo "${VERSION_CODENAME}")
cat << SRC | sudo tee /etc/apt/sources.list.d/intel-graphics.list >/dev/null
deb [arch=$(dpkg --print-architecture)\
 signed-by=/etc/apt/keyrings/intel-graphics.asc]\
 https://repositories.intel.com/gpu/ubuntu ${DISTRIBUTION} unified
SRC

cat <<'EXE' | sudo bash
apt-get update;echo
apt-get upgrade -y;echo
apt-get install -y intel-gpu-tools
apt-get install -y intel-opencl-icd libze1 libze-intel-gpu1 clinfo intel-media-va-driver-non-free
apt-get clean
EXE

ollama ps
ls /dev/dri
sudo intel_gpu_top
clinfo | grep 'Device Name'

cat <<'EXE' | sudo bash
rm /etc/apt/sources.list.d/intel-graphics.list
rm /etc/apt/keyrings/intel-graphics.asc
EXE
Ollama » Config 
cat <<'INI' | sudo tee /etc/systemd/system/ollama.service >/dev/null
[Unit]
Description=Ollama Service
After=network-online.target

[Service]
User=ollama
Group=ollama

# Optimizations for Yoga Pro 7i (16GB Target)
Environment="OLLAMA_KV_CACHE_TYPE=q4_0"
Environment="OLLAMA_FLASH_ATTENTION=1"
Environment="OLLAMA_NUM_PARALLEL=1"
Environment="OLLAMA_NUM_CTX=16384"
Environment="OLLAMA_KEEP_ALIVE=5m"
Environment="OLLAMA_NUM_GPU=999"

# Intel optimizations if available
# Environment="OLLAMA_INTEL_GPU=1"

Environment="PATH=/home/shahed/.nvm/versions/node/v18.20.7/bin:/home/shahed/.pyenv/plugins/pyenv-virtualenv/shims:/home/shahed/.pyenv/shims:/home/shahed/.pyenv/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/snap/bin:/snap/bin:/home/shahed/.local/share/JetBrains/Toolbox/scripts"

ExecStart=/usr/local/bin/ollama serve
Restart=always
RestartSec=3

[Install]
WantedBy=default.target
INI

cat <<'EXE' | sudo bash
systemctl daemon-reload
systemctl restart ollama.service
systemctl status  ollama.service
EXE
Ollama » Model 
ollama pull gpt-oss:20b
:'
pulling manifest 
pulling e7b273f96360: 100% ▕███████████████████████████████████████████▏  13 GB
pulling fa6710a93d78: 100% ▕███████████████████████████████████████████▏ 7.2 KB
pulling f60356777647: 100% ▕███████████████████████████████████████████▏  11 KB
pulling d8ba2f9a17b3: 100% ▕███████████████████████████████████████████▏   18 B
pulling 776beb3adb23: 100% ▕███████████████████████████████████████████▏  489 B
verifying sha256 digest 
writing manifest 
success 
'
Claude » Install 
curl -fsSL https://claude.ai/install.sh  | bash
:'
Setting up Claude Code...

✔ Claude Code successfully installed!
  Version: 2.1.63
  Location: ~/.local/bin/claude
  Next: Run claude --help to get started

⚠ Setup notes:
  • Native installation exists but ~/.local/bin is not in your PATH. Run:
  echo 'export PATH="$HOME/.local/bin:$PATH"' >> ~/.bashrc && source ~/.bashrc
✅ Installation complete!
'

cat <<'ENV' | tee -a ${HOME}/.bashrc >/dev/null

# claude config
export PATH="$HOME/.local/bin:$PATH" >> ~/.bashrc
ENV
source ~/.bashrc && sleep 1
claude --version
Claude » Config 
export ANTHROPIC_BASE_URL=http://localhost:11434
export ANTHROPIC_AUTH_TOKEN=ollama
export ANTHROPIC_API_KEY=''

claude --model gpt-oss:20b
ollama stop    gpt-oss:20b
ollama ps

References

References

Retrieved from "https://wiki.chorke.org/index.php?title=AI/Ollama&oldid=16704"