Text + image to text models, tools and agents

332 in the catalogue today. Every one with what it costs, who sells it and where it stands. 332 in all, a hundred to a page; this is page 3 of 4.

A picture and a question in, an answer in words out. This is what makes a model useful on a screenshot, a chart, a scanned page or a photograph of a whiteboard, and it is now the default shape for a capable general model.

• Ovis2.5-9BATH-MaaS
• Ovis2.6-30B-A3BATH-MaaS
• OvisOCR2ATH-MaaS
• PaddleOCR-VLpaddlepaddle$0.02→$0.02per Mtok in / out1 selling
• PaddleOCR-VL-1.5paddlepaddle
• PaddleOCR-VL-1.6paddlepaddle
• Phi-3.5-vision-instructMicrosoft1st of 26· 2 boards$0.13→$0.52per Mtok in / out1 selling
• Phi-4-reasoning-vision-15BMicrosoft
• Pixtral 12B 2409Mistral AI$0.15→$0.15per Mtok in / out3 selling
• Pixtral Large (25.02)Mistral AI$2→$6per Mtok in / out1 selling
• Qianfan-OCRBaidu$0.067→$0.27per Mtok in / out1 selling
• QvQ Max$1.2→$4.8per Mtok in / out1 selling
• Qwen 3.8 Flash NextAlibaba9th of 128· 2 boards
• Qwen 3.8 MaxAlibaba1st of 28· 24 boards$1.65→$4.95per Mtok in / out9 selling
• Qwen-Drive-1.0-4BAlibaba
• Qwen-VL MaxAlibaba41st of 50· 2 boards$0.8→$3.2per Mtok in / out2 selling
• Qwen-VL OCRAlibaba$0.72→$0.72per Mtok in / out1 selling
• Qwen-VL PlusAlibaba$0.21→$0.63per Mtok in / out2 selling
• Qwen2-VL-2B-InstructAlibaba$0.1→$0.1per Mtok in / out1 selling
• Qwen2-VL-7B-InstructAlibaba24th of 50· 2 boards$0.2→$0.2per Mtok in / out1 selling
• Qwen2.5 VL 72B InstructAlibaba4th of 50· 5 boards$0.8→$0.8per Mtok in / out7 selling
• Qwen2.5-VL 7BAlibaba$0.35→$1.05per Mtok in / out1 selling
• Qwen2.5-VL-3B-InstructAlibaba$0.2→$0.2per Mtok in / out1 selling
• Qwen2.5-VL-7B-InstructAlibaba$0.2→$0.2per Mtok in / out1 selling
• Qwen3 VL 235B A22B InstructAlibaba$0.2→$0.88per Mtok in / out10 selling
• Qwen3 VL 30B A3B InstructAlibaba$0.13→$0.52per Mtok in / out8 selling
• Qwen3 VL 30B A3B ThinkingAlibaba$0.15→$0.6per Mtok in / out7 selling
• Qwen3 VL 32B InstructAlibaba$0.1→$0.42per Mtok in / out5 selling
• Qwen3 VL 8B InstructAlibaba$0.08→$0.5per Mtok in / out8 selling
• Qwen3 VL 8B ThinkingAlibaba$0.18→$2.09per Mtok in / out3 selling
• Qwen3-VL 235BAlibaba$0.21→$1.9per Mtok in / out1 selling
• Qwen3-VL 30B-A3BAlibaba$0.2→$0.8per Mtok in / out1 selling
• Qwen3-VL PlusAlibaba$0.2→$1.6per Mtok in / out1 selling
• Qwen3-VL-2B-InstructAlibaba
• Qwen3-VL-32BAlibaba
• Qwen3.5 27B BlueStar v3 DerestrictedAlibaba$0.6→$1.2per Mtok in / out1 selling
• Qwen3.5 27B Queen DerestrictedAlibaba$0.6→$1.2per Mtok in / out1 selling
• Qwen3.5-27B-Claude-4.6-Opus-Reasoning-DistilledJackrong
• Qwen3.5-35B-A3B-BaseAlibaba$0.54→$1.33per Mtok in / out, training lane1 selling
• Qwen3.5-397B-A17B-TEEAlibaba$0.45→$3per Mtok in / out1 selling
• Qwen3.5-40B-Claude-4.6-Opus-Deckard-Heretic-Uncensored-ThinkingDavidAU
• Qwen3.5-4BAlibaba162nd of 302· 2 boards$0.03→$0.15per Mtok in / out2 selling
• Qwen3.5-9B-BaseAlibaba$0.66→$2per Mtok in / out, training lane1 selling
• Qwen3.5-9B-Claude-4.6-HighIQ-THINKING-HERETIC-UNCENSOREDDavidAU
• Qwen3.6-27B-TEEAlibaba$0.3→$2per Mtok in / out1 selling
• Qwen3.8-2.4T-A95BAlibaba31st of 302· 3 boards$2→$6per Mtok in / out9 selling
• Qwen3.8-27B-TEEAlibaba$0.32→$2.5per Mtok in / out1 selling
• Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NM-DAUDavidAU
• Qwen3.8-27B-UncensoredOrcaRouter
• Qwythos-27B-v1empero-ai
• R-4BYannQi
• Rio-3.5-Open-397BPrefeitura do Rio de Janeiro (City of Rio de Janeiro)
• RolmOCRReducto$0.2→$0.2per Mtok in / out1 selling
• Seed 1.6ByteDance22nd of 262$0.12→$0.3per Mtok in / out4 selling
• Seed-1.8$0.12→$0.3per Mtok in / out4 selling
• SmolVLM2-2.2B-InstructHugging Face Smol Models Research
• SonarPerplexity34th of 54$1→$1per Mtok in / out, low lane3 selling
• Sonar ProPerplexity31st of 34$3→$15per Mtok in / out, medium lane3 selling
• Sonar Pro SearchPerplexity$3→$15per Mtok in / out, medium lane2 selling
• Sonar Reasoning ProPerplexity29th of 34$2→$8per Mtok in / out, low lane3 selling
• Step 3.5 FlashStepFun3rd of 61· 2 boards$0.1→$0.31per Mtok in / out8 selling
• Step 3.7 FlashStepFun123rd of 302· 14 boards$0.2→$1.2per Mtok in / out9 selling
• Step3-VL-10BStepFun226th of 302
• Swift-Qwen3.8-27bukisai
• Thomson-1.0-SmallThomson Reuters
• UI-TARS 7BByteDance$0.1→$0.2per Mtok in / out1 selling
• UI-TARS-72B-DPOByteDance Seed
• UI-TARS-7B-DPOByteDance Seed
• UI-TARS-7B-SFTByteDance Seed
• Unlimited-OCRBaidu
• Venice Role Play Uncensored$0.5→$2per Mtok in / out1 selling
• Venice Uncensored 1.2$0.2→$0.9per Mtok in / out1 selling
• Yi-VL-34B01-ai
• Youtu-VL-4B-InstructTencent Hunyuan
• aya-vision-32bCohere Labs
• aya-vision-8bCohere Labs
• blip-vqa-baseSalesforce
• blip2-opt-2.7bSalesforce5th of 26
• bu-30b-a3b-previewBrowser Use
• chameleon-7bfacebook
• chandraDatalab
• chandra-ocr-2Datalab
• deepseek-vl-7b-chatdeepseek-ai
• deepseek-vl2DeepSeek
• deepseek-vl2-smalldeepseek-ai
• deepseek-vl2-tinydeepseek-ai
• deplotGoogle
• diffusiongemma-26B-A4B-itGoogle
• dots.mocrdots studio
• dots.ocrdots studio
• dots3-note-prevdots studio
• fuyu-8bAdept AI Labs
• gemma-3-12b-it-qat-q4_0-unquantizedGoogle
• gemma-3-27b-it-abliteratedmlabonne
• gemma-3-4b-ptGoogle
• gemma4Google$0.18→$0.5per Mtok in / out1 selling
• granite-docling-258MIBM Granite
• grok-code-fast-1xAI152nd of 202· 5 boards$5→$25per Mtok in / out3 selling
• idefics2-8bHuggingFaceM4
• internlm-xcomposer2d5-7binternlm