github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-06 08:02:14 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #15449] Ollama CLI should ask which model to load in case user has not chosen yet. #9874

New Issue
Open
opened 2026-04-12 22:44:13 -05:00 by GiteaMirror · 8 comments
GiteaMirror commented 2026-04-12 22:44:13 -05:00
Owner
Copy Link

Originally created by @lipstick-turtleback on GitHub (Apr 9, 2026).
Original GitHub issue: https://github.com/ollama/ollama/issues/15449

➜  ~ ollama list
NAME                      ID              SIZE      MODIFIED
x/flux2-klein:4b          50a0c0ab15ac    5.7 GB    4 hours ago
gemma4:31b-cloud          c382fbfbc73b    -         14 hours ago
gemma4:latest             c6eb396dbd59    9.6 GB    44 hours ago
gemma4:e2b                7fbdbf8f5e45    7.2 GB    6 days ago
gemma3:27b-cloud          9e1580299085    -         5 weeks ago
glm-5:cloud               c313cd065935    -         6 weeks ago
lfm2.5-thinking:latest    95bd9d45385f    731 MB    2 months ago
➜  ~ ollama run
Error: requires at least 1 arg(s), only received 0

Why ollama cant just provide the list of avaialble models to run and ask which one to load?
this will save a lot of time for cli users :)

Image
Originally created by @lipstick-turtleback on GitHub (Apr 9, 2026). Original GitHub issue: https://github.com/ollama/ollama/issues/15449 ``` ➜ ~ ollama list NAME ID SIZE MODIFIED x/flux2-klein:4b 50a0c0ab15ac 5.7 GB 4 hours ago gemma4:31b-cloud c382fbfbc73b - 14 hours ago gemma4:latest c6eb396dbd59 9.6 GB 44 hours ago gemma4:e2b 7fbdbf8f5e45 7.2 GB 6 days ago gemma3:27b-cloud 9e1580299085 - 5 weeks ago glm-5:cloud c313cd065935 - 6 weeks ago lfm2.5-thinking:latest 95bd9d45385f 731 MB 2 months ago ➜ ~ ollama run Error: requires at least 1 arg(s), only received 0 ``` Why ollama cant just provide the list of avaialble models to run and ask which one to load? this will save a lot of time for cli users :) <img width="472" height="206" alt="Image" src="https://github.com/user-attachments/assets/260c7aac-4ae8-41f0-890e-7cc35f6aa544" />
GiteaMirror added the feature request label 2026-04-12 22:44:13 -05:00
GiteaMirror commented 2026-04-12 22:44:14 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Apr 9, 2026):

#5380

<!-- gh-comment-id:4213671553 --> @rick-github commented on GitHub (Apr 9, 2026): #5380
GiteaMirror commented 2026-04-12 22:44:15 -05:00
Author
Owner
Copy Link

@somera commented on GitHub (Apr 9, 2026):

#5380

Here extended version of the script. All embeddings will be excluded.

#!/usr/bin/env bash
set -euo pipefail

OLLAMA_HOST="${OLLAMA_HOST:-http://localhost:11434}"

require_cmd() {
  command -v "$1" >/dev/null 2>&1 || {
    printf 'Error: required command not found: %s\n' "$1" >&2
    exit 127
  }
}

require_cmd curl
require_cmd jq
require_cmd ollama
require_cmd sort

api_get() {
  curl -fsS "${OLLAMA_HOST%/}$1"
}

api_post() {
  local endpoint="$1"
  local data="$2"
  curl -fsS \
    -H 'Content-Type: application/json' \
    -d "$data" \
    "${OLLAMA_HOST%/}$endpoint"
}

is_selectable_model() {
  local model="$1"
  local show_json

  if ! show_json="$(api_post '/api/show' "$(jq -nc --arg model "$model" '{model: $model}')" 2>/dev/null)"; then
    return 1
  fi

  jq -e '
    (.capabilities // []) | index("completion")
  ' >/dev/null <<<"$show_json"
}

# Enable colors only for terminals and only if NO_COLOR is not set.
if [[ -t 1 && -z "${NO_COLOR:-}" ]]; then
  CLOUD_COLOR=$'\033[1;36m'   # bold cyan
  TAG_COLOR=$'\033[2;36m'     # dim cyan
  RESET_COLOR=$'\033[0m'
else
  CLOUD_COLOR=''
  TAG_COLOR=''
  RESET_COLOR=''
fi

tmp_json="$(mktemp)"
trap 'rm -f "$tmp_json"' EXIT

if ! api_get '/api/tags' >"$tmp_json"; then
  printf 'Error: unable to query the Ollama API at %s.\n' "$OLLAMA_HOST" >&2
  exit 1
fi

declare -a all_models=()
declare -a models=()
declare -A model_is_cloud=()

while IFS=$'\t' read -r model_name is_cloud; do
  [[ -n "$model_name" ]] || continue
  all_models+=("$model_name")
  model_is_cloud["$model_name"]="$is_cloud"
done < <(
  jq -r '
    .models[]
    | [
        .name,
        (if ((.remote_host // "") != "" or (.remote_model // "") != "") then "1" else "0" end)
      ]
    | @tsv
  ' "$tmp_json"
)

if ((${#all_models[@]} == 0)); then
  printf 'Error: no models found.\n' >&2
  exit 1
fi

for model in "${all_models[@]}"; do
  if is_selectable_model "$model"; then
    models+=("$model")
  fi
done

if ((${#models[@]} == 0)); then
  printf 'Error: no selectable chat/completion models found.\n' >&2
  exit 1
fi

mapfile -t models < <(
  printf '%s\n' "${models[@]}" | sort -f
)

printf 'Available models:\n'
for i in "${!models[@]}"; do
  model="${models[$i]}"
  if [[ "${model_is_cloud[$model]:-0}" == "1" ]]; then
    printf '[%2d] %s%s%s %s[cloud]%s\n' \
      "$i" \
      "$CLOUD_COLOR" "$model" "$RESET_COLOR" \
      "$TAG_COLOR" "$RESET_COLOR"
  else
    printf '[%2d] %s\n' "$i" "$model"
  fi
done

while true; do
  read -r -p 'Select model [q=quit]: ' selection || {
    printf '\n'
    exit 0
  }

  case "$selection" in
    q|Q|quit|exit)
      exit 0
      ;;
  esac

  if [[ "$selection" =~ ^[0-9]+$ ]] && (( selection < ${#models[@]} )); then
    break
  fi

  printf 'Invalid selection. Please enter a number between 0 and %d or q to quit.\n' "$(( ${#models[@]} - 1 ))" >&2
done

exec ollama run "${models[$selection]}"
<!-- gh-comment-id:4218177446 --> @somera commented on GitHub (Apr 9, 2026): > [#5380](https://github.com/ollama/ollama/issues/5380) Here extended version of the script. All embeddings will be excluded. ``` #!/usr/bin/env bash set -euo pipefail OLLAMA_HOST="${OLLAMA_HOST:-http://localhost:11434}" require_cmd() { command -v "$1" >/dev/null 2>&1 || { printf 'Error: required command not found: %s\n' "$1" >&2 exit 127 } } require_cmd curl require_cmd jq require_cmd ollama require_cmd sort api_get() { curl -fsS "${OLLAMA_HOST%/}$1" } api_post() { local endpoint="$1" local data="$2" curl -fsS \ -H 'Content-Type: application/json' \ -d "$data" \ "${OLLAMA_HOST%/}$endpoint" } is_selectable_model() { local model="$1" local show_json if ! show_json="$(api_post '/api/show' "$(jq -nc --arg model "$model" '{model: $model}')" 2>/dev/null)"; then return 1 fi jq -e ' (.capabilities // []) | index("completion") ' >/dev/null <<<"$show_json" } # Enable colors only for terminals and only if NO_COLOR is not set. if [[ -t 1 && -z "${NO_COLOR:-}" ]]; then CLOUD_COLOR=$'\033[1;36m' # bold cyan TAG_COLOR=$'\033[2;36m' # dim cyan RESET_COLOR=$'\033[0m' else CLOUD_COLOR='' TAG_COLOR='' RESET_COLOR='' fi tmp_json="$(mktemp)" trap 'rm -f "$tmp_json"' EXIT if ! api_get '/api/tags' >"$tmp_json"; then printf 'Error: unable to query the Ollama API at %s.\n' "$OLLAMA_HOST" >&2 exit 1 fi declare -a all_models=() declare -a models=() declare -A model_is_cloud=() while IFS=$'\t' read -r model_name is_cloud; do [[ -n "$model_name" ]] || continue all_models+=("$model_name") model_is_cloud["$model_name"]="$is_cloud" done < <( jq -r ' .models[] | [ .name, (if ((.remote_host // "") != "" or (.remote_model // "") != "") then "1" else "0" end) ] | @tsv ' "$tmp_json" ) if ((${#all_models[@]} == 0)); then printf 'Error: no models found.\n' >&2 exit 1 fi for model in "${all_models[@]}"; do if is_selectable_model "$model"; then models+=("$model") fi done if ((${#models[@]} == 0)); then printf 'Error: no selectable chat/completion models found.\n' >&2 exit 1 fi mapfile -t models < <( printf '%s\n' "${models[@]}" | sort -f ) printf 'Available models:\n' for i in "${!models[@]}"; do model="${models[$i]}" if [[ "${model_is_cloud[$model]:-0}" == "1" ]]; then printf '[%2d] %s%s%s %s[cloud]%s\n' \ "$i" \ "$CLOUD_COLOR" "$model" "$RESET_COLOR" \ "$TAG_COLOR" "$RESET_COLOR" else printf '[%2d] %s\n' "$i" "$model" fi done while true; do read -r -p 'Select model [q=quit]: ' selection || { printf '\n' exit 0 } case "$selection" in q|Q|quit|exit) exit 0 ;; esac if [[ "$selection" =~ ^[0-9]+$ ]] && (( selection < ${#models[@]} )); then break fi printf 'Invalid selection. Please enter a number between 0 and %d or q to quit.\n' "$(( ${#models[@]} - 1 ))" >&2 done exec ollama run "${models[$selection]}" ```
GiteaMirror commented 2026-04-12 22:44:16 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Apr 9, 2026):

Embedding models can be run from the CLI and can be useful for generating one-off embeddings.

<!-- gh-comment-id:4218192979 --> @rick-github commented on GitHub (Apr 9, 2026): Embedding models can be run from the CLI and can be useful for generating one-off embeddings.
GiteaMirror commented 2026-04-12 22:44:16 -05:00
Author
Owner
Copy Link

@somera commented on GitHub (Apr 9, 2026):

I know. But in this case you need the text:
ollama run mxbai-embed-large:latest "test"

If you run it without text you get an error:

$ ollama run mxbai-embed-large:latest
Error: embedding models require input text. Usage: ollama run mxbai-embed-large:latest "your text here"
<!-- gh-comment-id:4218205640 --> @somera commented on GitHub (Apr 9, 2026): I know. But in this case you need the text: `ollama run mxbai-embed-large:latest "test"` If you run it without text you get an error: ``` $ ollama run mxbai-embed-large:latest Error: embedding models require input text. Usage: ollama run mxbai-embed-large:latest "your text here" ```
GiteaMirror commented 2026-04-12 22:44:17 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Apr 9, 2026):

You are correct, I thought I had previously used it interactively but I must be thinking of a different tool.

<!-- gh-comment-id:4218217607 --> @rick-github commented on GitHub (Apr 9, 2026): You are correct, I thought I had previously used it interactively but I must be thinking of a different tool.
GiteaMirror commented 2026-04-12 22:44:17 -05:00
Author
Owner
Copy Link

@somera commented on GitHub (Apr 9, 2026):

I updated my version above. The output is now:

Image
<!-- gh-comment-id:4218258697 --> @somera commented on GitHub (Apr 9, 2026): I updated my version above. The output is now: <img width="373" height="688" alt="Image" src="https://github.com/user-attachments/assets/0a9f4a3e-be87-43a3-a052-476fdeb93346" />
GiteaMirror commented 2026-04-12 22:44:19 -05:00
Author
Owner
Copy Link

@somera commented on GitHub (Apr 9, 2026):

@rick-github thx for the idea!

<!-- gh-comment-id:4218271066 --> @somera commented on GitHub (Apr 9, 2026): @rick-github thx for the idea!
GiteaMirror commented 2026-04-12 22:44:19 -05:00
Author
Owner
Copy Link

@drifkin commented on GitHub (Apr 9, 2026):

Good suggestion, we'll think about improving that. In the meantime, if you run ollama with no arguments, you can choose Chat with a model and press right arrow to see a model list.

<!-- gh-comment-id:4218369936 --> @drifkin commented on GitHub (Apr 9, 2026): Good suggestion, we'll think about improving that. In the meantime, if you run `ollama` with no arguments, you can choose `Chat with a model` and press right arrow to see a model list.
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 feature request
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#9874
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?