Best AI model for image understanding — November 2026
No single model is clearly best for image understanding on BenchLM's evidence. On AA-MMMU-Pro, run by Artificial Analysis, Claude Opus 5.5 (87.7%) and GPT-6 Astra (86.9%) are within 1.5 points of each other. The test is exam-style questions about one image. It does not establish reading of your own screenshots, handwriting or photos; try a few of your own images.
The shortlist for image understanding
Constraints on this page: a builder choosing by accuracy, hosted processing allowed, any price, ordinary input size. Change any of them under Refine. A small score gap does not establish a reliably better model.
Compare Claude Opus 5.5 vs GPT-6 AstraAll AA-MMMU-Pro results
01Claude Opus 5.5Tied at the top
Anthropic · Proprietary
87.7%
AA-MMMU-Pro
- Highest AA-MMMU-Pro result, with 1 other model within 1.5 points
- $8.00 for the stated workload
- 1M context
- Anthropic lists the model ID on each documented platform. Access details · checked September 22, 2026.
- Provider API offer: Claude API. Check which models qualify and whether payment details are required. Limits and source.
02GPT-6 AstraTied at the top
OpenAI · Proprietary
86.9%
AA-MMMU-Pro
- 86.9% on AA-MMMU-Pro, run by Artificial Analysis
- $20.00 for the stated workload
- 1.05M context
- OpenAI says GPT-6 Astra is rolling out first to enterprises in its Trusted Access Program, with API, ChatGPT plan, Azure, and Bedrock access following in the coming days. Enterprise access is off by default; eligible API customers can use Zero Data Retention. Access details · checked September 3, 2026.
- Provider API offer: OpenAI API, Free usage tier. Check which models qualify and whether payment details are required. Limits and source.
Model page and evidenceCompare with Claude Opus 5.5Open benchmark source
03GPT-6.1 Sol
OpenAI · Proprietary
86.0%
AA-MMMU-Pro
- 86.0% on AA-MMMU-Pro, run by Artificial Analysis
- $4.00 for the stated workload
- 1.05M context
- OpenAI lists GPT-6.1 Sol as an API model. Use Responses for tool calling; Chat Completions supports the model without tool calling. Access details · checked September 29, 2026.
- Provider API offer: OpenAI API, Free usage tier. Check which models qualify and whether payment details are required. Limits and source.
Model page and evidenceCompare with Claude Opus 5.5Open benchmark source
04GPT-6.1 Terra
OpenAI · Proprietary
84.7%
AA-MMMU-Pro
- 84.7% on AA-MMMU-Pro, run by Artificial Analysis
- $2.00 for the stated workload
- 1.05M context
- OpenAI launched gpt-6.1-terra in the API and in ChatGPT Work and Codex for Plus, Pro, Enterprise, and Edu users. Use Responses for tool calling; Chat Completions supports the model without tool calling. Access details · checked October 26, 2026.
- Provider API offer: OpenAI API, Free usage tier. Check which models qualify and whether payment details are required. Limits and source.
Model page and evidenceCompare with Claude Opus 5.5Open benchmark source
05Kimi K3.1 Pro
Moonshot AI · Open Weight
83.9%
AA-MMMU-Pro
- 83.9% on AA-MMMU-Pro, run by Artificial Analysis
- $4.00 for the stated workload
- 1.05M context
- Open weights; check hardware before treating local use as feasible
- API access is not verified. Check the provider documentation before relying on it.
Model page and evidenceCompare with Claude Opus 5.5Open benchmark source
ChatGPT, Claude, Gemini, or Copilot?
No app is clearly ahead here. On AA-MMMU-Pro, ChatGPT’s GPT-6 Astra (86.9%) and Claude Opus 5.5 (87.7%) are within 1.5 points of each other.
- ChatGPTGPT-6 Astra scores 86.9% (highest of the three)Plus, $20 a monthPlan lists: GPT-6 Astra, GPT-6.1 Sol, GPT-6.1 Terra, GPT-6 Sol and GPT-6 Luna in Codex; GPT-6.1 Sol is available in Work and Codex, not Chat.
- ClaudeClaude Opus 5.5 scores 87.7% (highest of the three)Pro, $20 a monthPlan lists: Opus, Sonnet and Haiku; Fable is listed under usage credits rather than the included allowance.
- GeminiGemini 3.8 Flash scores 85.6%Google AI Pro, $19.99 a monthPlan lists: Gemini 4 Argon, the Pro model and the Flash model.
- Microsoft CopilotNot scored. Microsoft doesn’t name the model Copilot runs, so no benchmark result can be attached to it. How it compares on plans and data termsMicrosoft 365 Premium, $19.99 a month
Each entry is the highest-scoring model that company sells, not the model your plan opens by default, so check what the plan lists before you pay. The plan shown is each app’s plan nearest $20, read October 16, 2026 for ChatGPT, November 10, 2026 for Copilot, and November 11, 2026 for Claude and Gemini.
ChatGPT vs ClaudeChatGPT vs GeminiClaude vs GeminiChatGPT vs Copilot
The top ten for image understanding, with API prices
The first ten models on the same evidence and constraints as the shortlist, with each provider's API price in US dollars per million tokens. GPT-6 Luna costs the least of the ten for a workload of 1M input and 200k output tokens: $0.10 per million input tokens and $0.50 per million output tokens. Three of the ten have no API price on file.
| # | Model | AA-MMMU-Pro | Input | Output |
|---|---|---|---|---|
| 01 | Claude Opus 5.5Anthropic | 87.7% | $4.00 | $20.00 |
| 02 | GPT-6 AstraOpenAI | 86.9% | $10.00 | $50.00 |
| 03 | GPT-6.1 SolOpenAI | 86.0% | $2.00 | $10.00 |
| 04 | GPT-6.1 TerraOpenAI | 84.7% | $1.00 | $5.00 |
| 05 | Kimi K3.1 ProMoonshot AI | 83.9% | $2.00 | $10.00 |
| 06 | Inkling 2 SmallThinking Machines Lab | 81.6% | $0.30 | $1.00 |
| 07 | Muse Spark OpenMeta | 80.9% | No API price on file | |
| 08 | Qwen3.7 PlusAlibaba | 80.5% | No API price on file | |
| 09 | GPT-6 LunaOpenAI | 79.7% | $0.10 | $0.50 |
| 10 | Apodex 1.1Apodex | 79.2% | No API price on file | |
Prices are list API rates before caching, batch discounts and special tiers. Compare API pricing for every model.
Refine for your situation
Each link opens the selector with one answer changed. The address carries the answers, so your version is as shareable as this page.
- Change any answer in the selector
- Must run on my own hardware
- Cost matters most
- Speed matters most
- Long documents or a large codebase
- Choosing an everyday assistant, not an API
- Describe your own job instead
We keep what you type here, drafts included, for 180 days, and after that only the job and requirements read from it. Nothing is tied to your IP address or account. Leave out names and other personal details. Privacy
What to verify before choosing
- The test is exam-style questions about one image. It does not establish reading of your own screenshots, handwriting or photos; try a few of your own images.
- Ranked on one benchmark: AA-MMMU-Pro, run by Artificial Analysis. 114 models have a published result; a model without one is not listed.
Try three representative examples of your own work. Compare errors, time, cost, and the tools available in your actual setup.
What this shortlist rests on
One benchmark: AA-MMMU-Pro. College-level questions across about thirty subjects that need the image to answer: diagrams, charts, tables, maps and photographs. The Pro version drops questions a model could answer from the text alone. A result a model's own provider ran is left out, and a model nobody has run there is not listed.
- AA-MMMU-ProSets the order. Run by Artificial Analysis. 114 models have a published result.
- MMMU-ProRelated evidence, not used for the order. The same test as each lab reports it, on its own settings.
- RealWorldQARelated evidence, not used for the order. Questions about everyday photographs. Each lab reports its own run.
Questions
Which AI model is best for image understanding?
No single model is clearly best for image understanding on BenchLM's evidence. On AA-MMMU-Pro, run by Artificial Analysis, Claude Opus 5.5 (87.7%) and GPT-6 Astra (86.9%) are within 1.5 points of each other. The test is exam-style questions about one image. It does not establish reading of your own screenshots, handwriting or photos; try a few of your own images.
Which models come next for image understanding?
GPT-6.1 Sol (86.0%), GPT-6.1 Terra (84.7%), Kimi K3.1 Pro (83.9%) follow on the same benchmark. Compare the leaders on three representative examples of your own work.
How does BenchLM pick the best ai model for image understanding?
The page runs the LLM Selector with fixed answers: a builder choosing by accuracy, hosted processing allowed, any price, ordinary input size. The selector uses AA-MMMU-Pro, one benchmark run by Artificial Analysis on every listed model, filters by the stated constraints, and orders by that evidence. It never adds a hidden fit score or a bonus for open weights or reasoning style.
Why does this page use AA-MMMU-Pro and not a general score?
A general score answers several jobs at once. AA-MMMU-Pro is closer to this job. College-level questions across about thirty subjects that need the image to answer: diagrams, charts, tables, maps and photographs. The Pro version drops questions a model could answer from the text alone. Only results Artificial Analysis published itself are used, so a model nobody has run there is not listed. 114 models have a result.
What do the top models for image understanding cost?
Among the ten highest-ranked models on this page, GPT-6 Luna costs the least for a workload of 1M input and 200k output tokens: $0.10 per million input tokens and $0.50 per million output tokens. GPT-6 Astra costs the most, at $10.00 and $50.00. These are API rates in the November 13, 2026 data. Assistant subscriptions are priced separately.
Can I change the constraints?
Yes. Every link under "Refine" opens the selector with one answer changed, and the address carries the answers so a result can be shared or reopened against the current dataset.
Method: bench-align-v5.8-2026-09-30. Read the methodology and benchmark confidence pages for how scores and verification statuses are produced.
Watch the image understanding shortlist
One weekly email when rank, price, or benchmark evidence changes make this shortlist worth revisiting.
Read a sample issueJoin 7,000+ subscribers.