[DOCS] CLI ImageEdit and ImageVariation tools

[inference-gateway-maintainer]

Documentation Request

What needs documenting

The CLI (inference-gateway/cli) now supports editing an existing image and requesting variations of one, alongside image generation. Two new agent tools were added:

• ImageEdit - reads an image from a local file path and sends it with a prompt to POST /v1/images/edits using the configured image model (tools.image_edit.model, default openai/gpt-image-2). Parameters: image (required, local file path), prompt (required), quality (optional, auto default, low, medium, high, standard), size (optional, 1024x1024 default, 1536x1024, 1024x1536).
• ImageVariation - reads an image from a local file path and sends it to POST /v1/images/variations using the configured image model (tools.image_variation.model). Parameters: image (required, local file path), size (optional, same enum as above).

Both save the returned image under .infer/tmp/ and are config-driven (enabled, model, require_approval under tools.image_edit / tools.image_variation). They are independent of the model selected for the chat session, like the existing ImageGeneration tool.

Affected pages

• CLI tools reference (agent tools: ImageGeneration, ImageEdit, ImageVariation) if the docs site mirrors docs/tools-reference.md
• CLI shortcuts guide / media tools section if mirrored from docs/shortcuts-guide.md

Links

• CLI PR: feat(tools): add ImageEdit and ImageVariation tools cli#986 (resolves [FEATURE] Implement images/edits and images/variations support cli#985)
• CLI docs already updated in-repo: docs/tools-reference.md, docs/shortcuts-guide.md
• Gateway endpoints: POST /v1/images/edits and POST /v1/images/variations (feat: implement images edits and variations endpoints inference-gateway#536)

[edenreich]

@infer Can you work on this? use /ponytail:ponytail

[inference-gateway-maintainer]

Model: ollama_cloud/deepseek-v4-flash · View Job

Todos

• Add issue to project board and set status to In progress
• Create working branch and push it
• Read current cli.md and understand what needs to change
• Read CLI PR #986 docs changes to understand exact content
• Update cli.md: add Image row to Tool Categories table
• Update cli.md: add ImageEdit and ImageVariation tool docs
• Update cli.md: update shortcuts section for image tools
• Validate changes (lint, format, build)
• Commit, push, and create PR

Pull Request

#510

✅ Infer Result: Success

Done. Here's what changed:

cli.md - 3 additions:

1. Tool Categories table - Added an "Image" row listing ImageGeneration, ImageEdit, ImageVariation
2. ImageEdit Tool section - Documents the tool that reads a local image and sends it with a prompt to POST /v1/images/edits using tools.image_edit.model (default openai/gpt-image-2). Parameters: image (required), prompt (required), quality (optional: auto/low/medium/high/standard), size (optional: 1024x1024/1536x1024/1024x1536)
3. ImageVariation Tool section - Documents the tool that reads a local image and sends it to POST /v1/images/variations using tools.image_variation.model. Parameters: image (required), size (optional, same enum)

Both save results under .infer/tmp/ and are independent of the chat session model, matching the existing ImageGeneration pattern.

PR: #510 (draft, on fix/issue-509)

Model: ollama_cloud/deepseek-v4-flash · Exit Code: 0 · Duration: 3m 27s

Tokens: 6,307,173 in · 11,973 out · 6,319,146 total (46 requests)

Tool calls: 52 total · 90% success rate

Traces

Session: 1785633483-cd61043d

session (standard, success)                                    3m25s
├── chat ollama_cloud/deepseek-v4-flash                         2.9s
│   ╰── POST /v1/chat/completions [inference-gateway]           2.9s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     2.9s
├── execute_tool TodoWrite call_5h3gvc4l                        84µs
├── chat ollama_cloud/deepseek-v4-flash                         1.9s
│   ╰── POST /v1/chat/completions [inference-gateway]           1.9s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     1.9s
├── execute_tool Read call_b1sudvbo                            132µs [error: tool_error]
├── execute_tool Bash call_cexizvcp                            255ms
├── execute_tool Read call_uz01j8fj                            112µs [error: tool_error]
├── chat ollama_cloud/deepseek-v4-flash                         1.2s
│   ╰── POST /v1/chat/completions [inference-gateway]           1.2s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     1.2s
├── chat ollama_cloud/deepseek-v4-flash                         1.3s
│   ╰── POST /v1/chat/completions [inference-gateway]           1.3s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     1.3s
├── execute_tool Tree call_114a0cin                              2ms
├── chat ollama_cloud/deepseek-v4-flash                         1.2s
│   ╰── POST /v1/chat/completions [inference-gateway]           1.2s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     1.2s
├── chat ollama_cloud/deepseek-v4-flash                         1.5s
│   ╰── POST /v1/chat/completions [inference-gateway]           1.5s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     1.5s
├── execute_tool Tree call_sbpcwr7g                              2ms
├── chat ollama_cloud/deepseek-v4-flash                         1.5s
│   ╰── POST /v1/chat/completions [inference-gateway]           1.5s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     1.5s
├── execute_tool Grep call_8dlzex8a                             1.5s
├── chat ollama_cloud/deepseek-v4-flash                         3.3s
│   ╰── POST /v1/chat/completions [inference-gateway]           3.3s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     3.3s
├── execute_tool Grep call_9je47y0p                            853ms
├── chat ollama_cloud/deepseek-v4-flash                         1.7s
│   ╰── POST /v1/chat/completions [inference-gateway]           1.7s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     1.7s
├── execute_tool Read call_ui23fgb3                            288µs
├── chat ollama_cloud/deepseek-v4-flash                         2.5s
│   ╰── POST /v1/chat/completions [inference-gateway]           2.5s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     2.5s
├── execute_tool TodoWrite call_tjmtv87j                        77µs
├── chat ollama_cloud/deepseek-v4-flash                         1.9s
│   ╰── POST /v1/chat/completions [inference-gateway]           1.9s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     1.9s
├── execute_tool Bash call_lpnnlhef                             1.1s
├── chat ollama_cloud/deepseek-v4-flash                         1.4s
│   ╰── POST /v1/chat/completions [inference-gateway]           1.4s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     1.4s
├── execute_tool Bash call_3ez8g7a3                            511ms
├── chat ollama_cloud/deepseek-v4-flash                         2.7s
│   ╰── POST /v1/chat/completions [inference-gateway]           2.7s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     2.7s
├── execute_tool TodoWrite call_r4vwvhsv                        76µs
├── chat ollama_cloud/deepseek-v4-flash                         1.6s
│   ╰── POST /v1/chat/completions [inference-gateway]           1.6s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     1.5s
├── execute_tool Read call_pk22twce                            286µs
├── execute_tool Bash call_v8ggf95d                             56ms
├── execute_tool Read call_a3zrbqxn                              2ms
├── chat ollama_cloud/deepseek-v4-flash                         4.0s
│   ╰── POST /v1/chat/completions [inference-gateway]           3.9s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     3.9s
├── execute_tool Bash call_kum5wcdy                            288ms
├── execute_tool Bash call_0wnd6e90                             1.1s
├── chat ollama_cloud/deepseek-v4-flash                         3.0s
│   ╰── POST /v1/chat/completions [inference-gateway]           3.0s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     3.0s
├── execute_tool TodoWrite call_ojlgz0o8                        73µs
├── chat ollama_cloud/deepseek-v4-flash                         2.5s
│   ╰── POST /v1/chat/completions [inference-gateway]           2.5s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     2.5s
├── chat ollama_cloud/deepseek-v4-flash                         3.0s
│   ╰── POST /v1/chat/completions [inference-gateway]           2.9s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     2.9s
├── execute_tool Bash call_qlpx68kz                             35ms [error: tool_error]
├── chat ollama_cloud/deepseek-v4-flash                         2.0s
│   ╰── POST /v1/chat/completions [inference-gateway]           2.0s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     2.0s
├── chat ollama_cloud/deepseek-v4-flash                         1.9s
│   ╰── POST /v1/chat/completions [inference-gateway]           1.9s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     1.8s
├── execute_tool Bash call_xssv5ow4                            672ms
├── chat ollama_cloud/deepseek-v4-flash                         4.8s
│   ╰── POST /v1/chat/completions [inference-gateway]           4.8s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     4.8s
├── execute_tool Grep call_w2cjx9px                              5ms
├── chat ollama_cloud/deepseek-v4-flash                         2.6s
│   ╰── POST /v1/chat/completions [inference-gateway]           2.6s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     2.5s
├── execute_tool Read call_43eh05uq                            621µs
├── chat ollama_cloud/deepseek-v4-flash                         6.1s
│   ╰── POST /v1/chat/completions [inference-gateway]           6.1s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     6.1s
├── chat ollama_cloud/deepseek-v4-flash                         2.3s
│   ╰── POST /v1/chat/completions [inference-gateway]           2.3s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     2.3s
├── execute_tool Read call_765ole6k                            181µs
├── chat ollama_cloud/deepseek-v4-flash                         6.4s
│   ╰── POST /v1/chat/completions [inference-gateway]           6.4s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     6.3s
├── execute_tool TodoWrite call_ktqvjdrc                        92µs
├── chat ollama_cloud/deepseek-v4-flash                         6.3s
│   ╰── POST /v1/chat/completions [inference-gateway]           6.3s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     6.3s
├── execute_tool Edit call_r21mcrxp                            438µs [error: tool_error]
├── chat ollama_cloud/deepseek-v4-flash                         2.0s
│   ╰── POST /v1/chat/completions [inference-gateway]           2.0s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     2.0s
├── execute_tool Read call_ek5msw11                            181µs
├── chat ollama_cloud/deepseek-v4-flash                         3.6s
│   ╰── POST /v1/chat/completions [inference-gateway]           3.6s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     3.6s
├── execute_tool Edit call_a563b1k6                              2ms
├── chat ollama_cloud/deepseek-v4-flash                         8.3s
│   ╰── POST /v1/chat/completions [inference-gateway]           8.3s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     8.3s
├── execute_tool TodoWrite call_76jgnpj0                        78µs
├── chat ollama_cloud/deepseek-v4-flash                         4.8s
│   ╰── POST /v1/chat/completions [inference-gateway]           4.7s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     4.7s
├── execute_tool Read call_puyniq61                            205µs
├── chat ollama_cloud/deepseek-v4-flash                         7.0s
│   ╰── POST /v1/chat/completions [inference-gateway]           7.0s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     7.0s
├── execute_tool Edit call_j8vy9351                              2ms
├── chat ollama_cloud/deepseek-v4-flash                         9.1s
│   ╰── POST /v1/chat/completions [inference-gateway]           9.0s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     9.0s
├── execute_tool TodoWrite call_5q1a8idn                       271µs
├── chat ollama_cloud/deepseek-v4-flash                         6.4s
│   ╰── POST /v1/chat/completions [inference-gateway]           6.3s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     6.3s
├── execute_tool Bash call_iayek9ep                             3.7s [error: tool_error]
├── chat ollama_cloud/deepseek-v4-flash                         3.6s
│   ╰── POST /v1/chat/completions [inference-gateway]           3.6s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     3.5s
├── execute_tool Bash call_ep3vwkyw                             3.9s
├── chat ollama_cloud/deepseek-v4-flash                         3.5s
│   ╰── POST /v1/chat/completions [inference-gateway]           3.5s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     3.5s
├── execute_tool Bash call_gt3lyeru                             1.5s
├── chat ollama_cloud/deepseek-v4-flash                         5.9s
│   ╰── POST /v1/chat/completions [inference-gateway]           5.9s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     5.9s
├── execute_tool Bash call_5cl4qyfq                            20.8s
├── chat ollama_cloud/deepseek-v4-flash                         4.3s
│   ╰── POST /v1/chat/completions [inference-gateway]           4.3s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     4.3s
├── execute_tool TodoWrite call_m9f8ofen                        91µs
├── chat ollama_cloud/deepseek-v4-flash                         4.2s
│   ╰── POST /v1/chat/completions [inference-gateway]           4.1s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     4.1s
├── execute_tool Write call_4ofue00y                           673µs
├── chat ollama_cloud/deepseek-v4-flash                         3.1s
│   ╰── POST /v1/chat/completions [inference-gateway]           3.1s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     3.1s
├── execute_tool Bash call_svdgk0yo                              7ms
├── chat ollama_cloud/deepseek-v4-flash                         3.1s
│   ╰── POST /v1/chat/completions [inference-gateway]           3.1s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     3.1s
├── execute_tool Bash call_vaed9ahx                             11ms
├── chat ollama_cloud/deepseek-v4-flash                         3.2s
│   ╰── POST /v1/chat/completions [inference-gateway]           3.2s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     3.2s
├── execute_tool Bash call_w6j04iv6                             1.1s
├── chat ollama_cloud/deepseek-v4-flash                         3.0s
│   ╰── POST /v1/chat/completions [inference-gateway]           3.0s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     2.9s
├── execute_tool Bash call_bm7ip99r                             1.6s
├── chat ollama_cloud/deepseek-v4-flash                         3.5s
│   ╰── POST /v1/chat/completions [inference-gateway]           3.5s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     3.4s
├── execute_tool Bash call_e5j4giwp                            536ms
├── chat ollama_cloud/deepseek-v4-flash                         3.5s
│   ╰── POST /v1/chat/completions [inference-gateway]           3.5s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     3.5s
├── execute_tool Bash call_v3v87dxy                            300ms
├── execute_tool Bash call_x532chkx                              6ms
├── execute_tool Bash call_w7ag7hsb                              6ms
├── chat ollama_cloud/deepseek-v4-flash                         4.0s
│   ╰── POST /v1/chat/completions [inference-gateway]           3.9s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     3.9s
├── execute_tool TodoWrite call_juz9whow                        72µs
├── chat ollama_cloud/deepseek-v4-flash                         3.9s
│   ╰── POST /v1/chat/completions [inference-gateway]           3.9s
│       ╰── POST /proxy/:provider/*path [inference-gateway]     3.9s
╰── chat ollama_cloud/deepseek-v4-flash                         3.1s

Stats

Tool Calls

╭───────────┬───────┬───────┬────────╮
│ Tool      │ Calls │ Fail% │ Avg    │
├───────────┼───────┼───────┼────────┤
│ Bash      │ 20    │ 10%   │ 1868ms │
│ Read      │ 9     │ 22%   │ 0ms    │
│ TodoWrite │ 9     │ 0%    │ 0ms    │
│ Edit      │ 3     │ 33%   │ 1ms    │
│ Grep      │ 3     │ 0%    │ 777ms  │
│ Tree      │ 2     │ 0%    │ 2ms    │
│ Write     │ 1     │ 0%    │ 0ms    │
╰───────────┴───────┴───────┴────────╯

Token Usage

╭────────────────────────────────┬─────────┬────────┬────────────┬─────────┬──────╮
│ Model                          │ Prompt  │ Cached │ Completion │ Total   │ Cost │
├────────────────────────────────┼─────────┼────────┼────────────┼─────────┼──────┤
│ ollama_cloud/deepseek-v4-flash │ 6582117 │ 0      │ 12047      │ 6594164 │ -    │
╰────────────────────────────────┴─────────┴────────┴────────────┴─────────┴──────╯

Sessions

╭───────────┬──────────┬──────────╮
│ Execution │ Mode     │ Sessions │
├───────────┼──────────┼──────────┤
│ headless  │ standard │ 1        │
╰───────────┴──────────┴──────────╯

⚠️ 5 log(s)

• Read call_uz01j8fj:

  NOT_FOUND: /home/runner/work/docs/docs/tools-reference.md
  

• Read call_b1sudvbo:

  NOT_FOUND: /home/runner/work/docs/docs/shortcuts-guide.md
  

• Bash call_qlpx68kz:

  exit status 1: accepts at most 1 arg(s), received 2
  

• Edit call_r21mcrxp:

  old_string not found in file /home/runner/work/docs/docs/cli.md
  
  Possible matches found:
  
  Near line 669:
  | **Computer Use**      | GetLatestScreenshot, MouseMove, MouseClick, MouseScroll, KeyboardType, GetFocusedApp, ActivateApp | GUI automation - see the Computer Use section above                                                                  |
  | **MCP**               | `MCP_<server>_<tool>`                                                                             | Dynamically registered tools from MCP servers - see [MCP](/mcp/)                                                     |
  
  Hint: Ensure exact whitespace and indentation match. Use the Read tool to see the current file content.
  

• Bash call_iayek9ep:

  exit status 1: $ prettier --check .
  Checking formatting...
  [�[33mwarn�[39m] cli.md
  [�[33mwarn�[39m] Code style issues found in the above file. Run Prettier with --write to fix.
  error: script "format:check" exited with code 1
  

Triggered by edenreich · Infer Action · OSS