Vision Model Leaderboard | LMArena Rankings

Read LMArena rankings for AI models that understand images. Your assistant compares the returned names, ranks, scores and votes; no image or extra details are needed.

Your AI

Use with your AI

For tasks with several steps, continue in the AI you already use.

CodexClaude CodeCursorWindsurfCline

View instructions

Use Toolmu get-vision-leaderboard-lmarena to read the Vision Arena rankings for models that understand images. Reuse a working Toolmu connection, or follow https://toolmu.com/SKILL.md to connect. Check current requirements, availability and the cumulative Credits ceiling, then execute within my authorization and budget. The current contract requires no inputs and documents no named query controls; do not ask me for an image, prompt, model identifier or technical parameters. If I give an ordinary display count, select up to that many records after retrieval rather than inventing a source filter or count parameter; keep the real returned count if fewer are available. Keep complete model names and variants distinct, using source references internally without asking me to find them. Compare source ranks, scores, votes and links only when returned, and retain missing values, blank organizations and original rank/rating bounds without guessing a method. Add any model price or capacity information only as supplied, leaving unconfirmed units and service terms unknown and separate from this lookup’s charge. This reads rankings rather than analyzing images, running models or proving suitability, access or use rights. Bracketed example text is replaced by my own content.

Task:
Read LMArena rankings for AI models that understand images. Compare the returned models by complete name, rank, score, votes and available reference links. Explain missing details in plain language.

Paste into your connected AI's conversation.

Connection guide

Up to 0.012 credits / call

The final charge is based on confirmed usage. Unused reserved credits return to your available balance.

LMArena

Vision Model Leaderboard | LMArena Rankings

Read LMArena rankings for AI models that understand images. Your assistant compares the returned names, ranks, scores and votes; no image or extra details are needed.

You receive

A comparison of the returned image-understanding models, with complete names, source ranks, scores, votes and reference links when available. Your assistant keeps model versions separate and shows any missing details clearly.

Start with your task

Read LMArena rankings for AI models that understand images. Compare the returned models by complete name, rank, score, votes and available reference links. Explain missing details in plain language.

What to know

The tool reads model rankings. It does not analyze your image or run a model, and the ranking cannot guarantee how a model will perform on your material. Check the actual model service before making a cost or access decision.

Inputs & outputs

Inspect the current inputs, availability, and maximum charge before starting.

Input schema

Input schema
Field / Path	Type	Requirement	Meaning
Optional query values`queryParams`	`object`	Optional	Optional object for query values. No nested query keys are named, and image, prompt, model-selection, filter or top-N controls are not documented. Omit it when no optional query values are needed.

View raw JSON

{
  "type": "object",
  "required": [],
  "properties": {
    "queryParams": {
      "type": "object",
      "properties": {},
      "additionalProperties": true
    }
  },
  "additionalProperties": false
}

Result schema

No fixed output contract is available. Field types reflect saved result evidence; individual responses can differ.

Result schema
Field / Path	Type	Meaning
Vision Arena model records`result.data.entries`	`array`	Collection of model/configuration records returned by the vision leaderboard. Its size and coverage may change; retain the actual returned list rather than assuming a fixed count or exhaustive catalog.
Source rank`result.data.entries[].rank`	`number`	Source rank of this returned record. Read it within the Vision Arena comparison; it is not a guaranteed suitability score for your material.
Votes`result.data.entries[].votes`	`number`	Source vote count associated with the record. It does not establish a count of unique voters or a verified sampling method.
Source rating`result.data.entries[].rating`	`number`	Source rating for the model record. The score scale and calculation method are not established here; do not turn it into an accuracy percentage.
License label`result.data.entries[].license`	`string`	Source license label. It does not establish verified license terms, current access or permission for your intended use.
Model identifier`result.data.entries[].modelKey`	`string`	Source identifier for the returned model/configuration. Preserve it exactly; it is not a documented selection parameter for this lookup.
Model reference link`result.data.entries[].modelUrl`	`string`	Reference link associated with the model record. It is not an image output or a guarantee that the linked service is currently accessible.
Rank bound (rankLower)`result.data.entries[].rankLower`	`number`	Rank-bound value under the source’s lower label. Preserve the label and number even when it exceeds rankUpper; the method and confidence level are unknown.
Rank bound (rankUpper)`result.data.entries[].rankUpper`	`number`	Rank-bound value under the source’s upper label. Keep the original value rather than swapping it with rankLower; the method and confidence level are unknown.
Lower rating bound`result.data.entries[].ratingLower`	`number`	Lower rating-bound value supplied by the source. Preserve it without inferring a confidence level or statistical method.
Upper rating bound`result.data.entries[].ratingUpper`	`number`	Upper rating-bound value supplied by the source. Preserve it without inferring a confidence level or statistical method.
Release label`result.data.entries[].releaseType`	`null \| string`	Source release label, such as pre_release, or null. Its access conditions and precise meaning are not established; null remains unknown.
Context value`result.data.entries[].contextLength`	`number \| null`	Source context value or null. The unit and capacity meaning are not established; do not interpret it as a supported image count or resolution.
Complete model name`result.data.entries[].modelDisplayName`	`string`	Human-readable model/configuration name supplied by the source. Keep the complete name so distinct variants are not silently merged.
Organization`result.data.entries[].modelOrganization`	`string`	Organization string supplied for the record. An empty string is not a verified organization; do not fill it from the model name or URL.
Input price per million`result.data.entries[].inputPricePerMillion`	`number \| null`	Source per-million input price value or null. Currency, counted quantity and billing conditions are not established; a missing value is unknown.
Output price per million`result.data.entries[].outputPricePerMillion`	`number \| null`	Source per-million output price value or null. Currency, counted quantity and billing conditions are not established; a missing value is unknown.

Additional notes

Compare complete configurations before combining similarly named entries. A shortened family name can conceal the variant represented by a record; keep both the display name and source identifier when sharing a shortlist. The identifier distinguishes a returned entry and does not add a model-selection control to this leaderboard lookup.

Use scores to guide what you investigate, rather than to predict a result on your own material. Read the vote count with the rating, preserve returned bounds, and keep their source labels even when rank bounds run in the opposite numerical order. The values do not establish a confidence level or a verified voting method.

Use attached context and price information to identify questions for the model’s actual service. A context number does not establish an image count or supported resolution, and per-million price labels do not establish the counted quantity or currency. Keep missing values unknown, then verify the service’s current terms before making a workload or cost decision.

Questions about this tool.

Inputs, results and details to know before running it.

Get connected

Similar names may represent different model versions. Your assistant keeps their complete names separate so the comparison stays clear; you do not need to find a model number.

Vision Model Leaderboard | LMArena Rankings

Vision Model Leaderboard | LMArena Rankings

You receive

Start with your task

What to know

Inputs & outputs

Input schema

Result schema

Ask for image-understanding model rankings

Choose how many models to show

Compare the actual returned models

Questions about this tool.

You receive

Start with your task

What to know

Inputs & outputs

Input schema

Result schema

Ask for image-understanding model rankings

Choose how many models to show

Compare the actual returned models

Questions about this tool.

Why keep the complete model name?

Does context length describe the images I can submit?

Are model price values the cost of this leaderboard lookup?