jaothan/DockerGenAI_Streamlit
0
1#syntax = docker/dockerfile:1.4
2
3FROM ollama/ollama:latest AS ollama
4FROM babashka/babashka:latest
5
6# just using as a client - never as a server
7COPY --from=ollama /bin/ollama ./bin/ollama
8
9COPY <<EOF pull_model.clj
10(ns pull-model
11 (:require [babashka.process :as process]
12 [clojure.core.async :as async]))
13
14(try
15 (let [llm (get (System/getenv) "LLM")
16 url (get (System/getenv) "OLLAMA_BASE_URL")]
17 (println (format "pulling ollama model %s using %s" llm url))
18 (if (and llm
19 url
20 (not (#{"gpt-4" "gpt-3.5" "claudev2" "gpt-4o" "gpt-4-turbo"} llm))
21 (not (some #(.startsWith llm %) ["ai21.jamba-instruct-v1:0"
22 "amazon.titan"
23 "anthropic.claude"
24 "cohere.command"
25 "meta.llama"
26 "mistral.mi"])))
27
28 ;; ----------------------------------------------------------------------
29 ;; just call `ollama pull` here - create OLLAMA_HOST from OLLAMA_BASE_URL
30 ;; ----------------------------------------------------------------------
31 ;; TODO - this still doesn't show progress properly when run from docker compose
32
33 (let [done (async/chan)]
34 (async/go-loop [n 0]
35 (let [[v _] (async/alts! [done (async/timeout 5000)])]
36 (if (= :stop v) :stopped (do (println (format "... pulling model (%ss) - will take several minutes" (* n 10))) (recur (inc n))))))
37 (process/shell {:env {"OLLAMA_HOST" url "HOME" (System/getProperty "user.home")} :out :inherit :err :inherit} (format "bash -c './bin/ollama show %s --modelfile > /dev/null || ./bin/ollama pull %s'" llm llm))
38 (async/>!! done :stop))
39
40 (println "OLLAMA model only pulled if both LLM and OLLAMA_BASE_URL are set and the LLM model is not gpt")))
41 (catch Throwable _ (System/exit 1)))
42EOF
43
44ENTRYPOINT ["bb", "-f", "pull_model.clj"]
45
46 