{"repo": "cheretbe/notes", "files": {"ai.md": {"3bdf00ab37fa6b8835e17cf641e641e480c366e6": {"t": 1785746380.8253977, "n": 2245, "h": "ea00bd8d70cc1782", "c": "### ollama\n* Browse models: https://ollama.com/library\n* https://github.com/ollama/ollama/blob/main/docs/api.md\n* Difference between between /api/generate and /api/chat: https://github.com/ollama/ollama/issues/2774\n```shell\n# as root 😅\n# [!!] check if it causes duplicate deb entries in /etc/apt/sources.list.d/contrib.list\n#      https://github.com/ollama/ollama/issues/6923\ncurl -fsSL https://ollama.com/install.sh | sh\n\nollama pull deepseek-coder:1.3b-base\nollama list\n# remotely\ncurl http://192.168.1.100:11434/api/tags | jq\nollama rm deepseek-coder:1.3b-base\n\n# 1b 7b etc. is the number of (billions of) parameters in the model\n# View this number for a downloaded model\nollama show deepseek-coder\ncurl http://192.168.1.100:11434/api/show -d '{\n  \"name\": \"deepseek-coder:latest\"\n}' | jq '.details'\n\n# Benchmarking CPU usage\n# View the number of cores available\nnproc\n# gpt-oss:20b always uses thinking mode\n# https://github.com/ollama/ollama/issues/11751\n# Compare response time by changing \"num_thread\" option\ncurl http://192.168.1.100:11434/api/generate -d '\n{  \n\"model\": \"deepseek-r1:1.5b\",  \n\"prompt\": \"Why is the blue sky blue?\",  \n\"stream\": false,\n\"think\": false,\n\"keep_alive\": 0,\n\"options\":{\n  \"temperature\": 0,\n  \"system\": \"You are a bored assistant. Provide short answers.\",\n  \"num_thread\": 20\n}\n}' | jq 'del(.context)'\n```\n### Continue plugin\nMinimal working config for VSCode (~/.continue/config.yaml)\n```yaml\nname: Local Agent\nversion: 1.0.0\nschema: v1\nmodels:\n  - name: Autodetect\n    provider: ollama\n    apiBase: \"http://192.168.1.100:11434\"\n    model: AUTODETECT\nroles:\n  - chat\n  - edit\n  - apply\n\ncontext:\n  - provider: code\n  - provider: docs\n  - provider: diff\n  - provider: terminal\n  - provider: problems\n  - provider: folder\n  - provider: codebase\n```\n### Deepseek\n* API: https://api-docs.deepseek.com/\n* https://platform.deepseek.com/api_keys\n\n```shell\n# Test API key\ncurl https://api.deepseek.com/v1/chat/completions \\\n  -H \"Authorization: Bearer ${DEEPSEEK_API_KEY}\" \\\n  -H \"Content-Type: application/json\" \\\n  -d '{\n    \"model\": \"deepseek-coder\",\n    \"messages\": [\n      {\"role\": \"system\", \"content\": \"You are a helpful assistant.\"},\n      {\"role\": \"user\", \"content\": \"Hello\"}\n    ],\n    \"max_tokens\": 50\n  }'\n```\n"}, "3db31ea4a2726c7972371c6641e954cd14aa2119": {"t": 1785763143.5800653, "n": 2245, "h": "ea00bd8d70cc1782", "c": "### ollama\n* Browse models: https://ollama.com/library\n* https://github.com/ollama/ollama/blob/main/docs/api.md\n* Difference between between /api/generate and /api/chat: https://github.com/ollama/ollama/issues/2774\n```shell\n# as root 😅\n# [!!] check if it causes duplicate deb entries in /etc/apt/sources.list.d/contrib.list\n#      https://github.com/ollama/ollama/issues/6923\ncurl -fsSL https://ollama.com/install.sh | sh\n\nollama pull deepseek-coder:1.3b-base\nollama list\n# remotely\ncurl http://192.168.1.100:11434/api/tags | jq\nollama rm deepseek-coder:1.3b-base\n\n# 1b 7b etc. is the number of (billions of) parameters in the model\n# View this number for a downloaded model\nollama show deepseek-coder\ncurl http://192.168.1.100:11434/api/show -d '{\n  \"name\": \"deepseek-coder:latest\"\n}' | jq '.details'\n\n# Benchmarking CPU usage\n# View the number of cores available\nnproc\n# gpt-oss:20b always uses thinking mode\n# https://github.com/ollama/ollama/issues/11751\n# Compare response time by changing \"num_thread\" option\ncurl http://192.168.1.100:11434/api/generate -d '\n{  \n\"model\": \"deepseek-r1:1.5b\",  \n\"prompt\": \"Why is the blue sky blue?\",  \n\"stream\": false,\n\"think\": false,\n\"keep_alive\": 0,\n\"options\":{\n  \"temperature\": 0,\n  \"system\": \"You are a bored assistant. Provide short answers.\",\n  \"num_thread\": 20\n}\n}' | jq 'del(.context)'\n```\n### Continue plugin\nMinimal working config for VSCode (~/.continue/config.yaml)\n```yaml\nname: Local Agent\nversion: 1.0.0\nschema: v1\nmodels:\n  - name: Autodetect\n    provider: ollama\n    apiBase: \"http://192.168.1.100:11434\"\n    model: AUTODETECT\nroles:\n  - chat\n  - edit\n  - apply\n\ncontext:\n  - provider: code\n  - provider: docs\n  - provider: diff\n  - provider: terminal\n  - provider: problems\n  - provider: folder\n  - provider: codebase\n```\n### Deepseek\n* API: https://api-docs.deepseek.com/\n* https://platform.deepseek.com/api_keys\n\n```shell\n# Test API key\ncurl https://api.deepseek.com/v1/chat/completions \\\n  -H \"Authorization: Bearer ${DEEPSEEK_API_KEY}\" \\\n  -H \"Content-Type: application/json\" \\\n  -d '{\n    \"model\": \"deepseek-coder\",\n    \"messages\": [\n      {\"role\": \"system\", \"content\": \"You are a helpful assistant.\"},\n      {\"role\": \"user\", \"content\": \"Hello\"}\n    ],\n    \"max_tokens\": 50\n  }'\n```\n"}, "f3a1e494cfd785788a8db5f68951f830805a7e8f": {"t": 1785763144.9571588, "n": 1871, "h": "69b56afcb358721e", "c": "### ollama\n* Browse models: https://ollama.com/library\n* https://github.com/ollama/ollama/blob/main/docs/api.md\n* Difference between between /api/generate and /api/chat: https://github.com/ollama/ollama/issues/2774\n```shell\n# as root 😅\n# [!!] check if it causes duplicate deb entries in /etc/apt/sources.list.d/contrib.list\n#      https://github.com/ollama/ollama/issues/6923\ncurl -fsSL https://ollama.com/install.sh | sh\n\nollama pull deepseek-coder:1.3b-base\nollama list\n# remotely\ncurl http://192.168.1.100:11434/api/tags | jq\nollama rm deepseek-coder:1.3b-base\n\n# 1b 7b etc. is the number of (billions of) parameters in the model\n# View this number for a downloaded model\nollama show deepseek-coder\ncurl http://192.168.1.100:11434/api/show -d '{\n  \"name\": \"deepseek-coder:latest\"\n}' | jq '.details'\n\n# Benchmarking CPU usage\n# View the number of cores available\nnproc\n# gpt-oss:20b always uses thinking mode\n# https://github.com/ollama/ollama/issues/11751\n# Compare response time by changing \"num_thread\" option\ncurl http://192.168.1.100:11434/api/generate -d '\n{  \n\"model\": \"deepseek-r1:1.5b\",  \n\"prompt\": \"Why is the blue sky blue?\",  \n\"stream\": false,\n\"think\": false,\n\"keep_alive\": 0,\n\"options\":{\n  \"temperature\": 0,\n  \"system\": \"You are a bored assistant. Provide short answers.\",\n  \"num_thread\": 20\n}\n}' | jq 'del(.context)'\n```\n### Continue plugin\nMinimal working config for VSCode (~/.continue/config.yaml)\n```yaml\nname: Local Agent\nversion: 1.0.0\nschema: v1\nmodels:\n  - name: Autodetect\n    provider: ollama\n    apiBase: \"http://192.168.1.100:11434\"\n    model: AUTODETECT\nroles:\n  - chat\n  - edit\n  - apply\n\ncontext:\n  - provider: code\n  - provider: docs\n  - provider: diff\n  - provider: terminal\n  - provider: problems\n  - provider: folder\n  - provider: codebase\n```\n### Deepseek\n* API: https://api-docs.deepseek.com/\n* https://platform.deepseek.com/api_keys\n"}, "7ea8f573b5afd894d8e620e0c5663de3d4c702de": {"t": 1785763146.5426714, "n": 1820, "h": "ebcefa8c7f682c10", "c": "### ollama\n* Browse models: https://ollama.com/library\n* https://github.com/ollama/ollama/blob/main/docs/api.md\n* Difference between between /api/generate and /api/chat: https://github.com/ollama/ollama/issues/2774\n```shell\n# as root 😅\n# [!!] check if it causes duplicate deb entries in /etc/apt/sources.list.d/contrib.list \ncurl -fsSL https://ollama.com/install.sh | sh\n\nollama pull deepseek-coder:1.3b-base\nollama list\n# remotely\ncurl http://192.168.1.100:11434/api/tags | jq\nollama rm deepseek-coder:1.3b-base\n\n# 1b 7b etc. is the number of (billions of) parameters in the model\n# View this number for a downloaded model\nollama show deepseek-coder\ncurl http://192.168.1.100:11434/api/show -d '{\n  \"name\": \"deepseek-coder:latest\"\n}' | jq '.details'\n\n# Benchmarking CPU usage\n# View the number of cores available\nnproc\n# gpt-oss:20b always uses thinking mode\n# https://github.com/ollama/ollama/issues/11751\n# Compare response time by changing \"num_thread\" option\ncurl http://192.168.1.100:11434/api/generate -d '\n{  \n\"model\": \"deepseek-r1:1.5b\",  \n\"prompt\": \"Why is the blue sky blue?\",  \n\"stream\": false,\n\"think\": false,\n\"keep_alive\": 0,\n\"options\":{\n  \"temperature\": 0,\n  \"system\": \"You are a bored assistant. Provide short answers.\",\n  \"num_thread\": 20\n}\n}' | jq 'del(.context)'\n```\n### Continue plugin\nMinimal working config for VSCode (~/.continue/config.yaml)\n```yaml\nname: Local Agent\nversion: 1.0.0\nschema: v1\nmodels:\n  - name: Autodetect\n    provider: ollama\n    apiBase: \"http://192.168.1.100:11434\"\n    model: AUTODETECT\nroles:\n  - chat\n  - edit\n  - apply\n\ncontext:\n  - provider: code\n  - provider: docs\n  - provider: diff\n  - provider: terminal\n  - provider: problems\n  - provider: folder\n  - provider: codebase\n```\n### Deepseek\n* API: https://api-docs.deepseek.com/\n* https://platform.deepseek.com/api_keys\n"}, "d97f048948ccb63906c8d733e8b1125721155580": {"t": 1785763147.7278748, "n": 1731, "h": "f7d2a1644b04fa72", "c": "### ollama\n* Browse models: https://ollama.com/library\n* https://github.com/ollama/ollama/blob/main/docs/api.md\n* Difference between between /api/generate and /api/chat: https://github.com/ollama/ollama/issues/2774\n```shell\n# as root 😅\ncurl -fsSL https://ollama.com/install.sh | sh\n\nollama pull deepseek-coder:1.3b-base\nollama list\n# remotely\ncurl http://192.168.1.100:11434/api/tags | jq\nollama rm deepseek-coder:1.3b-base\n\n# 1b 7b etc. is the number of (billions of) parameters in the model\n# View this number for a downloaded model\nollama show deepseek-coder\ncurl http://192.168.1.100:11434/api/show -d '{\n  \"name\": \"deepseek-coder:latest\"\n}' | jq '.details'\n\n# Benchmarking CPU usage\n# View the number of cores available\nnproc\n# gpt-oss:20b always uses thinking mode\n# https://github.com/ollama/ollama/issues/11751\n# Compare response time by changing \"num_thread\" option\ncurl http://192.168.1.100:11434/api/generate -d '\n{  \n\"model\": \"deepseek-r1:1.5b\",  \n\"prompt\": \"Why is the blue sky blue?\",  \n\"stream\": false,\n\"think\": false,\n\"keep_alive\": 0,\n\"options\":{\n  \"temperature\": 0,\n  \"system\": \"You are a bored assistant. Provide short answers.\",\n  \"num_thread\": 20\n}\n}' | jq 'del(.context)'\n```\n### Continue plugin\nMinimal working config for VSCode (~/.continue/config.yaml)\n```yaml\nname: Local Agent\nversion: 1.0.0\nschema: v1\nmodels:\n  - name: Autodetect\n    provider: ollama\n    apiBase: \"http://192.168.1.100:11434\"\n    model: AUTODETECT\nroles:\n  - chat\n  - edit\n  - apply\n\ncontext:\n  - provider: code\n  - provider: docs\n  - provider: diff\n  - provider: terminal\n  - provider: problems\n  - provider: folder\n  - provider: codebase\n```\n### Deepseek\n* API: https://api-docs.deepseek.com/\n* https://platform.deepseek.com/api_keys\n"}, "149e62995932f2140269f1127478e3fd4f43a623": {"t": 1785763150.0217688, "n": 1711, "h": "f5ad7aeb97c46de2", "c": "### ollama\n* Browse models: https://ollama.com/library\n* https://github.com/ollama/ollama/blob/main/docs/api.md\n* Difference between between /api/generate and /api/chat: https://github.com/ollama/ollama/issues/2774\n```shell\n# as root 😅\ncurl -fsSL https://ollama.com/install.sh | sh\n\nollama pull deepseek-coder:1.3b-base\nollama list\n# remotely\ncurl http://192.168.1.100:11434/api/tags | jq\nollama rm deepseek-coder:1.3b-base\n\n# 1b 7b etc. is the number of (billions of) parameters in the model\n# View this number for a downloaded model\nollama show deepseek-coder\ncurl http://192.168.1.100:11434/api/show -d '{\n  \"name\": \"deepseek-coder:latest\"\n}' | jq '.details'\n\n# Benchmarking CPU usage\n# View the number of cores available\nnproc\n# gpt-oss:20b always uses thinking mode\n# https://github.com/ollama/ollama/issues/11751\n# Compare response time by changing \"num_thread\" option\ncurl http://192.168.1.100:11434/api/generate -d '\n{  \n\"model\": \"deepseek-r1:1.5b\",  \n\"prompt\": \"Why is the blue sky blue?\",  \n\"stream\": false,\n\"think\": false,\n\"keep_alive\": 0,\n\"options\":{\n  \"temperature\": 0,\n  \"system\": \"You are a bored assistant. Provide short answers.\",\n  \"num_thread\": 20\n}\n}' | jq 'del(.context)'\n```\nMinimal working config for VSCode (~/.continue/config.yaml)\n```yaml\nname: Local Agent\nversion: 1.0.0\nschema: v1\nmodels:\n  - name: Autodetect\n    provider: ollama\n    apiBase: \"http://192.168.1.100:11434\"\n    model: AUTODETECT\nroles:\n  - chat\n  - edit\n  - apply\n\ncontext:\n  - provider: code\n  - provider: docs\n  - provider: diff\n  - provider: terminal\n  - provider: problems\n  - provider: folder\n  - provider: codebase\n```\n### Deepseek\n* API: https://api-docs.deepseek.com/\n* https://platform.deepseek.com/api_keys\n"}, "8444b026ed2069dca9781c3587efa8990fdc0465": {"t": 1785763151.5084145, "n": 1289, "h": "96229851bebdb565", "c": "### ollama\n* Browse models: https://ollama.com/library\n* https://github.com/ollama/ollama/blob/main/docs/api.md\n* Difference between between /api/generate and /api/chat: https://github.com/ollama/ollama/issues/2774\n```shell\n# as root 😅\ncurl -fsSL https://ollama.com/install.sh | sh\n\nollama pull deepseek-coder:1.3b-base\nollama list\n# remotely\ncurl http://192.168.1.100:11434/api/tags | jq\nollama rm deepseek-coder:1.3b-base\n\n# 1b 7b etc. is the number of (billions of) parameters in the model\n# View this number for a downloaded model\nollama show deepseek-coder\ncurl http://192.168.1.100:11434/api/show -d '{\n  \"name\": \"deepseek-coder:latest\"\n}' | jq '.details'\n\n# Benchmarking CPU usage\n# View the number of cores available\nnproc\n# gpt-oss:20b always uses thinking mode\n# https://github.com/ollama/ollama/issues/11751\n# Compare response time by changing \"num_thread\" option\ncurl http://192.168.1.100:11434/api/generate -d '\n{  \n\"model\": \"deepseek-r1:1.5b\",  \n\"prompt\": \"Why is the blue sky blue?\",  \n\"stream\": false,\n\"think\": false,\n\"keep_alive\": 0,\n\"options\":{\n  \"temperature\": 0,\n  \"system\": \"You are a bored assistant. Provide short answers.\",\n  \"num_thread\": 20\n}\n}' | jq 'del(.context)'\n```\n### Deepseek\n* API: https://api-docs.deepseek.com/\n* https://platform.deepseek.com/api_keys\n"}, "6c57b7cebf26198b4e82fd0bc6c282f98cea8901": {"t": 1785763152.9532712, "n": 1179, "h": "c6e064fffe6fca4a", "c": "### ollama\n* Browse models: https://ollama.com/library\n* https://github.com/ollama/ollama/blob/main/docs/api.md\n* Difference between between /api/generate and /api/chat: https://github.com/ollama/ollama/issues/2774\n```shell\n# as root 😅\ncurl -fsSL https://ollama.com/install.sh | sh\n\nollama pull deepseek-coder:1.3b-base\nollama list\n# remotely\ncurl http://192.168.1.100:11434/api/tags | jq\nollama rm deepseek-coder:1.3b-base\n\n# 1b 7b etc. is the number of (billions of) parameters in the model\n# View this number for a downloaded model\nollama show deepseek-coder\ncurl http://192.168.1.100:11434/api/show -d '{\n  \"name\": \"deepseek-coder:latest\"\n}' | jq '.details'\n\n# Benchmarking CPU usage\n# View the number of cores available\nnproc\n# Compare response time by changing \"num_thread\" option\ncurl http://192.168.1.100:11434/api/generate -d '\n{  \n\"model\": \"gpt-oss:20b\",  \n\"prompt\": \"Why is the blue sky blue?\",  \n\"stream\": false,\n\"think\": false,\n\"options\":{\n  \"temperature\": 0,\n  \"system\": \"You are a bored assistant. Provide short answers.\",\n  \"num_thread\": 20\n}\n}' | jq 'del(.context)'\n```\n### Deepseek\n* API: https://api-docs.deepseek.com/\n* https://platform.deepseek.com/api_keys\n"}, "b683a1b64ff0f120466eaa913d92bd0345bcbebd": {"t": 1785763154.3045635, "n": 1122, "h": "d5aa0ff075a9b797", "c": "### ollama\n* Browse models: https://ollama.com/library\n* Difference between between /api/generate and /api/chat: https://github.com/ollama/ollama/issues/2774\n```shell\n# as root 😅\ncurl -fsSL https://ollama.com/install.sh | sh\n\nollama pull deepseek-coder:1.3b-base\nollama list\n# remotely\ncurl http://192.168.1.100:11434/api/tags | jq\nollama rm deepseek-coder:1.3b-base\n\n# 1b 7b etc. is the number of (billions of) parameters in the model\n# View this number for a downloaded model\nollama show deepseek-coder\ncurl http://192.168.1.100:11434/api/show -d '{\n  \"name\": \"deepseek-coder:latest\"\n}' | jq '.details'\n\n# Benchmarking CPU usage\n# View the number of cores available\nnproc\n# Compare response time by changing \"num_thread\" option\ncurl http://192.168.1.100:11434/api/generate -d '\n{  \n\"model\": \"gpt-oss:20b\",  \n\"prompt\": \"Why is the blue sky blue?\",  \n\"stream\": false,\n\"think\": false,\n\"options\":{\n  \"temperature\": 0,\n  \"system\": \"You are a bored assistant. Provide short answers.\",\n  \"num_thread\": 20\n}\n}' | jq 'del(.context)'\n```\n### Deepseek\n* API: https://api-docs.deepseek.com/\n* https://platform.deepseek.com/api_keys\n"}, "70bb03fe9d327f4de626f7b8e99889cc0fc8b498": {"t": 1785763155.6486435, "n": 506, "h": "0921229b733e4e47", "c": "### ollama\n* Browse models: https://ollama.com/library\n```shell\n# as root 😅\ncurl -fsSL https://ollama.com/install.sh | sh\n\nollama pull deepseek-coder:1.3b-base\nollama list\n# remotely\ncurl http://192.168.1.100:11434/api/tags | jq\nollama rm deepseek-coder:1.3b-base\n\n# 1b 7b etc. is the number of (billions of) parameters in the model\n# View this number for a downloaded model\nollama show deepseek-coder\ncurl http://192.168.1.100:11434/api/show -d '{\n  \"name\": \"deepseek-coder:latest\"\n}' | jq '.details'\n```\n"}, "8ba20d676aae8bfc170a2f68f82fad976f17c662": {"t": 1785763157.0915926, "n": 211, "h": "a091bc3fae7289ba", "c": "### ollama\n* Browse models: https://ollama.com/library\n```shell\n# as root 😅\ncurl -fsSL https://ollama.com/install.sh | sh\n\nollama pull deepseek-coder:1.3b-base\nollama list\nollama rm deepseek-coder:1.3b-base\n```\n"}}}}