LLM Models · Multimodal
Browse the world's large language models. Compare parameters, benchmarks, VRAM and more.
Multimodal listings claim inputs beyond plain text, most often images, sometimes other files. Open this category when the job includes a screenshot, a scan, or a figure. Do not open it when the job is a diff. A vision-capable model is not automatically better at code, and a text-only flagship will not become multimodal because you paste an image into the wrong field.
Check the modality line against the about paragraph. Advertisements say "can see." The field says what the listing actually recorded. If they disagree, follow the field and tell us if the paragraph is a stale vendor blurb. Sources and corrections is how.
Multimodal does not decide open versus closed. A vision model can be weights you host or an API you pay for. Apply that cut first if your data cannot leave, then use this category. Open versus proprietary comes before the modality filter for that reason.
Some text-only chat models are paired, in the DeepSeek Harness world, with an inference plugin that describes images. That is a plugin, not a change to this row. If you need that adapter, read model inference plugins after you accept that the base model is blind.
Test with one image whose contents you can state in a sentence. If the model invents a control that is not in the picture, it is not ready for a UI review, regardless of the directory score. Keep the image small and non-confidential for the test.
Keep the test image in the ticket. When someone asks why this model was chosen, the image and the wrong-or-right caption are the evidence. A score is not. Delete the image if it was confidential. The note can describe the result without the pixels, and it should say whether the caption matched. A match you did not write down did not happen.
Claude Opus 5 is Anthropic’s flagship model for demanding reasoning, coding, and long-horizon agentic work. It is particularly strong at end-to-end software tasks, code review and bug finding, visual analysis...
Claude Opus 5 is Anthropic’s flagship model for demanding reasoning, coding, and long-horizon agentic work. It is particularly strong at end-to-end software tasks, code review and bug finding, visual analysis...
Fast-mode variant of [Opus 5](/anthropic/claude-opus-5) - identical capabilities with higher output speed at 2x pricing relative to regular Opus 5. Learn more in Anthropic's docs: https://platform.claude.com/docs/en/build-with-claude/fast-mode
This model always redirects to the latest model in the Claude Opus family.
This model always redirects to the latest model in the Google Gemini Flash family.
Gemini 3.7 Flash is a multimodal model from Google for fast agentic workflows, coding, and complex multi-step reasoning. It is designed for tasks that require responsive performance and reliable multi-step...
Gemini 3.7 Flash is a multimodal model from Google for fast agentic workflows, coding, and complex multi-step reasoning. It is designed for tasks that require responsive performance and reliable multi-step...
This model always redirects to the latest model in the Kimi family.
This model always redirects to the latest model in the Anthropic Claude Sonnet family.
Sonnet 5 is Anthropic's most capable Sonnet-class model, with frontier performance across coding, agents, and professional work. It supports adaptive thinking with selectable reasoning effort levels (low, medium, high, max,...
Sonnet 5 is Anthropic's most capable Sonnet-class model, with frontier performance across coding, agents, and professional work. It supports adaptive thinking with selectable reasoning effort levels (low, medium, high, max,...
Claude Fable 5 is a Mythos-class model from Anthropic, built for autonomous knowledge work and coding. It supports text, image, and file inputs with text output, with reasoning support and...