OCR Engines

Browse and compare 258 OCR engines across providers. Find the best engine for your use case.

Available Engines (174)

Gemini 3.1 Flash Lite

Benchmark: 60.7% benchSpeed: fast~821 pages / $1~$0.00122 per standard page

Mistral OCR 4.1

Mistral OCR 4.1 extracts page-wise Markdown with optional structured blocks and confidence scores.

200 pages / $1$0.005/page

Qwen: Qwen3.7 Flash

Qwen3.7 Flash is a vision-language reasoning model from Alibaba. It is suited for multimodal agents, visual coding, search, and computer in…

Speed: slow~7,910 pages / $1~$0.000126 per standard page

Qwen: Qwen3 VL 235B A22B Instruct

Qwen3-VL-235B-A22B Instruct is an open-weight multimodal model that unifies strong text generation with visual understanding across images…

Benchmark: 63% benchSpeed: medium~678 pages / $1~$0.00148 per standard page100 languages

Anthropic: Claude Haiku 5.5

Claude Haiku 5.5 is Anthropic's small, fast model for high-volume, cost-sensitive work such as summarization, subagents, and browser use.

~2,200 pages / $1~$0.000455 per standard page

NVIDIA: Nemotron 3.5 Content Safety

NVIDIA Nemotron 3.5 Content Safety is a compact 4B-parameter multimodal guardrail model from NVIDIA, fine-tuned from Google Gemma-3-4B. It…

~3,230 pages / $1~$0.000309 per standard page

Qwen: Qwen3.8 Omni Flash

Qwen3.8 Omni Flash is an omni-modal reasoning model from Alibaba and the first Qwen model built around agentic capabilities with native aud…

~2,050 pages / $1~$0.000487 per standard page

DeepSeek: DeepSeek V4 Flash Vision Exp

DeepSeek V4 Flash Vision Exp is an experimental vision-enabled version of DeepSeek V4 Flash 0731 from DeepSeek. It adds image understanding…

~1,270 pages / $1~$0.000789 per standard page

Z.ai: GLM 5.3 Flash

GLM-5.3-Flash is a native multimodal model from Z.ai. It is suited for efficient coding and long-horizon agent tasks. Its hybrid sparse and…

~1,420 pages / $1~$0.000703 per standard page

Perceptron: Perceptron Mk1.5

Perceptron Mk1.5 is Perceptron’s embodied reasoning model for physical agents. It accepts text, image, video, and audio input, and answers…

~895 pages / $1~$0.00112 per standard page

inclusionAI: Ling 3.0 Flash VL

Ling 3.0 Flash VL builds on Ling 3.0 Flash (124B total / 5.5B active MoE from InclusionAI), further strengthening its language capabilities…

~3,330 pages / $1~$0.000301 per standard page

Qwen3-VL 235B A22B Instruct (DeepInfra)

Qwen/Qwen3-VL-235B-A22B-Instruct — OCR-capable vision model.

Benchmark: 63% benchSpeed: slow~1,000 pages / $1~$0.000996 per standard page100 languages

GPT 4.1 Mini

GPT-4.1 Mini — Cost-effective OCR with GPT-4.1's improved accuracy.

Benchmark: 46.2% benchSpeed: medium~412 pages / $1~$0.00243 per standard page100 languages

DeepSeek: DeepSeek V4.1 Flash

DeepSeek V4.1 Flash is a sparse mixture-of-experts model from DeepSeek and the cost-efficient tier of the V4.1 family. DeepSeek reports tha…

~852 pages / $1~$0.00117 per standard page

Z.ai: GLM 5.3 FlashX

GLM-5.3-FlashX is the high-speed variant of Z.ai’s GLM-5.3-Flash, a native multimodal model delivering inference speeds of up to 200 tokens…

~571 pages / $1~$0.00175 per standard page

PrismML: Ternary Bonsai 2 27B

Bonsai 2 27B is a 27B-parameter reasoning model from PrismML derived from Qwen3.8-27B. It supports coding, mathematics, tool calling, and i…

~2,240 pages / $1~$0.000447 per standard page

Llama 3.2 11B Vision (DeepInfra)

meta-llama/Llama-3.2-11B-Vision-Instruct — OCR-capable vision model.

Speed: slow~1,170 pages / $1$0.000856/page100 languages

Mistral Large

mistral-large-2512 — OCR-capable vision model.

Benchmark: 58.5% benchSpeed: medium~477 pages / $1~$0.0021 per standard page

Mistral Medium

mistral-medium-2508 — OCR-capable vision model.

Benchmark: 61.9% benchSpeed: medium~602 pages / $1~$0.00166 per standard page100 languages

OpenAI: GPT-6 Luna Pro

GPT-6 Luna Pro is the same underlying model as GPT-6 Luna, served with `reasoning.mode` set to `pro` for higher-quality responses on comple…

~393 pages / $1~$0.00254 per standard page

OpenAI: GPT-6 Luna

GPT-6 Luna is the fast, cost-efficient model in OpenAI’s GPT-6 series, positioned below GPT-6 Sol. It is suited for high-volume and latency…

~1,130 pages / $1~$0.000881 per standard page

Mistral: Mistral Small 3.1 24B

Mistral Small 3.1 24B Instruct is an upgraded variant of Mistral Small 3 (2501), featuring 24 billion parameters with advanced multimodal c…

Speed: medium~1,510 pages / $1~$0.000663 per standard page100 languages

DeepSeek: DeepSeek Flash Latest

This model always redirects to the latest model in the DeepSeek Flash family.

~2,480 pages / $1~$0.000404 per standard page

Xiaomi: MiMo-V2.6-Flash

MiMo-V2.6-Flash is an open-source foundation model developed by Xiaomi. It is built on a Mixture-of-Experts architecture with 309B total pa…

~3,310 pages / $1~$0.000302 per standard page

Pareto 26.9

Pareto is a multimodal composite model built for research, coding, and agentic workflows, while delivering frontier-level performance acros…

~142 pages / $1~$0.00702 per standard page

ScaleDown: Summarize

Summarize is an abstractive summarization model from ScaleDown.

~28,900 pages / $1~$0.0000346 per standard page

Nano Banana 2.1

Nano Banana 2.1 appears to be a fictional or placeholder Google Gemini-based OCR engine slug that does not correspond to a known or publicl…

~175 pages / $1~$0.0057 per standard page

Meta: Muse Wonder

Muse Wonder is a frontier reasoning model from Meta, designed for complex personal and professional agentic tasks.

~60.9 pages / $1~$0.0164 per standard page

Google: Gemini 4 Argon

Gemini 4 Argon is Google's flagship reasoning model, built for complex software engineering, deep mathematical and scientific problem solvi…

~122 pages / $1~$0.00821 per standard page

Gemini 4 Argon

Gemini 4 Argon is Google's multimodal OCR engine leveraging the Gemini 4 model family to extract and interpret text across a wide range of…

~132 pages / $1~$0.0076 per standard page

Showing 30 of 174 available engines. Enable JavaScript to browse the full catalog with filters and sort.

Coming Soon (84)

Pareto

Pareto is a multimodal composite model built for research, coding, and agentic workflows, while delivering frontier-level performance acros…

~142 pages / $1~$0.00702 per standard page

ABBYY Cloud OCR

Enterprise-grade cloud OCR from ABBYY. Supports 200+ languages with high accuracy on printed text, handwriting, tables, and forms. ISO 2700…

~117 pages / $1$0.00856/page

Gemini 3 Flash Preview (Agent)

Price TBD

Anthropic: Claude Fable 5.1 (batch)

Claude Fable 5.1 improves on Claude Fable 5 across the board, with the biggest gains in agentic coding, long-running agentic workflows, and…

~48.7 pages / $1~$0.0205 per standard page

Gemini 2.5 Computer Use Preview 10-2025

Gemini 2.5 Computer Use Preview 10-2025 — OCR-capable vision model.

Price TBD

Mistral OCR

Mistral OCR — Dedicated OCR model by Mistral AI. Optimized for document text extraction.

Price TBD

Sakana: Namazu

Sakana Namazu is a Japanese-specialized reasoning model from Sakana AI, based on Kimi K2.6 with additional training for Japanese language a…

~293 pages / $1~$0.00341 per standard page

Anthropic: Claude 3.7 Sonnet

Claude 3.7 Sonnet is an advanced large language model with improved reasoning, coding, and problem-solving capabilities. It introduces a hy…

~173 pages / $1$0.005778/page

Gemini 2.5 Flash-Lite Preview Sep 2025

Gemini 2.5 Flash-Lite Preview Sep 2025 — OCR-capable vision model.

~3,120 pages / $1~$0.00032 per standard page

DeepSeek: DeepSeek V4.1 Flash (batch)

DeepSeek V4.1 Flash is a sparse mixture-of-experts model from DeepSeek, and the first built on the company's Causal Encoder-Decoder (CED) a…

~3,180 pages / $1~$0.000315 per standard page

Baidu: Qianfan-OCR-Fast

Qianfan-OCR-Fast is a domain-specific multimodal large model purpose-built for OCR. By leveraging specialized OCR training data while prese…

~904 pages / $1$0.001106/page

Anthropic: Claude Opus 4.7 (Fast)

Fast-mode variant of Opus 4.7 - identical capabilities with higher output speed at premium 6x pricing. Learn more in Anthropic's docs: htt…

~8.12 pages / $1~$0.123 per standard page

accounts/fireworks/routers/kimi-k2p7-code-fast

Price TBD

Ox Alpha

Ox Alpha is a reasoning model designed for coding, sustained agentic work, and production workloads. It is suited for long-horizon software…

Price TBD

Magistral Small

magistral-small-2509 — OCR-capable vision model.

~1,340 pages / $1$0.000749/page

OCR Engines

Browse and compare 258 OCR engines across providers. Find the best engine for your use case.

Available Engines (174)

Gemini 3.1 Flash Lite

Benchmark: 60.7% benchSpeed: fast~821 pages / $1~$0.00122 per standard page

Mistral OCR 4.1

Mistral OCR 4.1 extracts page-wise Markdown with optional structured blocks and confidence scores.

200 pages / $1$0.005/page

Qwen: Qwen3.7 Flash

Qwen3.7 Flash is a vision-language reasoning model from Alibaba. It is suited for multimodal agents, visual coding, search, and computer in…

Speed: slow~7,910 pages / $1~$0.000126 per standard page

Qwen: Qwen3 VL 235B A22B Instruct

Qwen3-VL-235B-A22B Instruct is an open-weight multimodal model that unifies strong text generation with visual understanding across images…

Benchmark: 63% benchSpeed: medium~678 pages / $1~$0.00148 per standard page100 languages

Anthropic: Claude Haiku 5.5

Claude Haiku 5.5 is Anthropic's small, fast model for high-volume, cost-sensitive work such as summarization, subagents, and browser use.

~2,200 pages / $1~$0.000455 per standard page

NVIDIA: Nemotron 3.5 Content Safety

NVIDIA Nemotron 3.5 Content Safety is a compact 4B-parameter multimodal guardrail model from NVIDIA, fine-tuned from Google Gemma-3-4B. It…

~3,230 pages / $1~$0.000309 per standard page

Qwen: Qwen3.8 Omni Flash

Qwen3.8 Omni Flash is an omni-modal reasoning model from Alibaba and the first Qwen model built around agentic capabilities with native aud…

~2,050 pages / $1~$0.000487 per standard page

DeepSeek: DeepSeek V4 Flash Vision Exp

DeepSeek V4 Flash Vision Exp is an experimental vision-enabled version of DeepSeek V4 Flash 0731 from DeepSeek. It adds image understanding…

~1,270 pages / $1~$0.000789 per standard page

Z.ai: GLM 5.3 Flash

GLM-5.3-Flash is a native multimodal model from Z.ai. It is suited for efficient coding and long-horizon agent tasks. Its hybrid sparse and…

~1,420 pages / $1~$0.000703 per standard page

Perceptron: Perceptron Mk1.5

Perceptron Mk1.5 is Perceptron’s embodied reasoning model for physical agents. It accepts text, image, video, and audio input, and answers…

~895 pages / $1~$0.00112 per standard page

inclusionAI: Ling 3.0 Flash VL

Ling 3.0 Flash VL builds on Ling 3.0 Flash (124B total / 5.5B active MoE from InclusionAI), further strengthening its language capabilities…

~3,330 pages / $1~$0.000301 per standard page

Qwen3-VL 235B A22B Instruct (DeepInfra)

Qwen/Qwen3-VL-235B-A22B-Instruct — OCR-capable vision model.

Benchmark: 63% benchSpeed: slow~1,000 pages / $1~$0.000996 per standard page100 languages

GPT 4.1 Mini

GPT-4.1 Mini — Cost-effective OCR with GPT-4.1's improved accuracy.

Benchmark: 46.2% benchSpeed: medium~412 pages / $1~$0.00243 per standard page100 languages

DeepSeek: DeepSeek V4.1 Flash

DeepSeek V4.1 Flash is a sparse mixture-of-experts model from DeepSeek and the cost-efficient tier of the V4.1 family. DeepSeek reports tha…

~852 pages / $1~$0.00117 per standard page

Z.ai: GLM 5.3 FlashX

GLM-5.3-FlashX is the high-speed variant of Z.ai’s GLM-5.3-Flash, a native multimodal model delivering inference speeds of up to 200 tokens…

~571 pages / $1~$0.00175 per standard page

PrismML: Ternary Bonsai 2 27B

Bonsai 2 27B is a 27B-parameter reasoning model from PrismML derived from Qwen3.8-27B. It supports coding, mathematics, tool calling, and i…

~2,240 pages / $1~$0.000447 per standard page

Llama 3.2 11B Vision (DeepInfra)

meta-llama/Llama-3.2-11B-Vision-Instruct — OCR-capable vision model.

Speed: slow~1,170 pages / $1$0.000856/page100 languages

Mistral Large

mistral-large-2512 — OCR-capable vision model.

Benchmark: 58.5% benchSpeed: medium~477 pages / $1~$0.0021 per standard page

Mistral Medium

mistral-medium-2508 — OCR-capable vision model.

Benchmark: 61.9% benchSpeed: medium~602 pages / $1~$0.00166 per standard page100 languages

OpenAI: GPT-6 Luna Pro

GPT-6 Luna Pro is the same underlying model as GPT-6 Luna, served with `reasoning.mode` set to `pro` for higher-quality responses on comple…

~393 pages / $1~$0.00254 per standard page

OpenAI: GPT-6 Luna

GPT-6 Luna is the fast, cost-efficient model in OpenAI’s GPT-6 series, positioned below GPT-6 Sol. It is suited for high-volume and latency…

~1,130 pages / $1~$0.000881 per standard page

Mistral: Mistral Small 3.1 24B

Mistral Small 3.1 24B Instruct is an upgraded variant of Mistral Small 3 (2501), featuring 24 billion parameters with advanced multimodal c…

Speed: medium~1,510 pages / $1~$0.000663 per standard page100 languages

DeepSeek: DeepSeek Flash Latest

This model always redirects to the latest model in the DeepSeek Flash family.

~2,480 pages / $1~$0.000404 per standard page

Xiaomi: MiMo-V2.6-Flash

MiMo-V2.6-Flash is an open-source foundation model developed by Xiaomi. It is built on a Mixture-of-Experts architecture with 309B total pa…

~3,310 pages / $1~$0.000302 per standard page

Pareto 26.9

Pareto is a multimodal composite model built for research, coding, and agentic workflows, while delivering frontier-level performance acros…

~142 pages / $1~$0.00702 per standard page

ScaleDown: Summarize

Summarize is an abstractive summarization model from ScaleDown.

~28,900 pages / $1~$0.0000346 per standard page

Nano Banana 2.1

Nano Banana 2.1 appears to be a fictional or placeholder Google Gemini-based OCR engine slug that does not correspond to a known or publicl…

~175 pages / $1~$0.0057 per standard page

Meta: Muse Wonder

Muse Wonder is a frontier reasoning model from Meta, designed for complex personal and professional agentic tasks.

~60.9 pages / $1~$0.0164 per standard page

Google: Gemini 4 Argon

Gemini 4 Argon is Google's flagship reasoning model, built for complex software engineering, deep mathematical and scientific problem solvi…

~122 pages / $1~$0.00821 per standard page

Gemini 4 Argon

Gemini 4 Argon is Google's multimodal OCR engine leveraging the Gemini 4 model family to extract and interpret text across a wide range of…

~132 pages / $1~$0.0076 per standard page

Showing 30 of 174 available engines. Enable JavaScript to browse the full catalog with filters and sort.

Engine	Provider	Benchmark	Speed	Price
Gemini 3.1 Flash Lite	Google Cloud	60.7%✓	1.2s	~821 pages / $1~$0.00122 per standard page
Mistral OCR 4.1	Mistral AI	No hard score	—	200 pages / $1$0.005/page
Qwen: Qwen3.7 Flash	OOpenRouter	No hard score	15.2s	~7,910 pages / $1~$0.000126 per standard page
Qwen: Qwen3 VL 235B A22B Instruct	OOpenRouter	63%✓	4.4s	~678 pages / $1~$0.00148 per standard page
Anthropic: Claude Haiku 5.5	OOpenRouter	No hard score	—	~2,200 pages / $1~$0.000455 per standard page
NVIDIA: Nemotron 3.5 Content Safety	OOpenRouter	No hard score	—	~3,230 pages / $1~$0.000309 per standard page
Qwen: Qwen3.8 Omni Flash	OOpenRouter	No hard score	—	~2,050 pages / $1~$0.000487 per standard page
Qwen3-VL 235B A22B Instruct (DeepInfra)	DeepInfra	63%✓	14.1s	~1,000 pages / $1~$0.000996 per standard page
GPT 4.1 Mini	OpenAI	46.2%✓	2.7s	~412 pages / $1~$0.00243 per standard page
DeepSeek: DeepSeek V4 Flash Vision Exp	OOpenRouter	No hard score	—	~1,270 pages / $1~$0.000789 per standard page
DeepSeek: DeepSeek V4.1 Flash	OOpenRouter	No hard score	—	~852 pages / $1~$0.00117 per standard page
inclusionAI: Ling 3.0 Flash VL	OOpenRouter	No hard score	—	~3,330 pages / $1~$0.000301 per standard page
Perceptron: Perceptron Mk1.5	OOpenRouter	No hard score	—	~895 pages / $1~$0.00112 per standard page
Z.ai: GLM 5.3 Flash	OOpenRouter	No hard score	—	~1,420 pages / $1~$0.000703 per standard page
Llama 3.2 11B Vision (DeepInfra)	DeepInfra	No hard score	22.3s	~1,170 pages / $1$0.000856/page
PrismML: Ternary Bonsai 2 27B	OOpenRouter	No hard score	—	~2,240 pages / $1~$0.000447 per standard page
Z.ai: GLM 5.3 FlashX	OOpenRouter	No hard score	—	~571 pages / $1~$0.00175 per standard page
Mistral Medium	Mistral AI	61.9%✓	2.0s	~602 pages / $1~$0.00166 per standard page
Mistral Large	Mistral AI	58.5%✓	4.8s	~477 pages / $1~$0.0021 per standard page
OpenAI: GPT-6 Luna Pro	OOpenRouter	No hard score	—	~393 pages / $1~$0.00254 per standard page
OpenAI: GPT-6 Luna	OOpenRouter	No hard score	—	~1,130 pages / $1~$0.000881 per standard page
Mistral: Mistral Small 3.1 24B	OOpenRouter	No hard score	1.7s	~1,510 pages / $1~$0.000663 per standard page
DeepSeek: DeepSeek Flash Latest	OOpenRouter	No hard score	—	~2,480 pages / $1~$0.000404 per standard page
Xiaomi: MiMo-V2.6-Flash	OOpenRouter	No hard score	—	~3,310 pages / $1~$0.000302 per standard page
Gemini 4 Argon	Google Cloud	No hard score	—	~132 pages / $1~$0.0076 per standard page
Google: Gemini 4 Argon	OOpenRouter	No hard score	—	~122 pages / $1~$0.00821 per standard page
Meta: Muse Wonder	OOpenRouter	No hard score	—	~60.9 pages / $1~$0.0164 per standard page
Nano Banana 2.1	Google Cloud	No hard score	—	~175 pages / $1~$0.0057 per standard page
Pareto 26.9	OOpenRouter	No hard score	—	~142 pages / $1~$0.00702 per standard page
ScaleDown: Summarize	OOpenRouter	No hard score	—	~28,900 pages / $1~$0.0000346 per standard page
Mistral Large 4	Mistral AI	No hard score	—	~556 pages / $1~$0.0018 per standard page
Xiaomi: MiMo-V2.6-Pro	OOpenRouter	No hard score	—	~1,060 pages / $1~$0.00094 per standard page
Cohere: Command A+	OOpenRouter	No hard score	—	~812 pages / $1~$0.00123 per standard page
StepFun: Step 5 Preview	OOpenRouter	No hard score	—	~383 pages / $1~$0.00261 per standard page
Datalab Convert (accurate)	DDatalab	77.2%✓	25.9s	100 pages / $1$0.01/page
AWS Textract	AWS	19.5%✓	450ms	~623 pages / $1$0.001605/page
Meta: Llama 4 Maverick	OOpenRouter	71%✓	4.9s	~1,560 pages / $1~$0.00064 per standard page
IBM Docling (hosted)	IIBM Docling (hosted)	No hard score	—	250 pages / $1$0.004/page
Anthropic: Claude Opus 4.5	OOpenRouter	67.5%✓	5.6s	~46.1 pages / $1~$0.0217 per standard page
Mistral: Mistral Medium 3	OOpenRouter	64.9%✓	3.9s	~684 pages / $1~$0.00146 per standard page
Mistral: Mistral Medium 3.1	OOpenRouter	64.6%✓	2.2s	~684 pages / $1~$0.00146 per standard page
Gemini 3 Flash Preview	Google Cloud	63.6%✓	7.2s	~184 pages / $1~$0.00543 per standard page
StepFun: Step 3.7 Flash	OOpenRouter	63.6%✓	11.9s	~1,150 pages / $1~$0.000867 per standard page
Anthropic: Claude Opus 4.6	OOpenRouter	63.4%✓	6.1s	~46.1 pages / $1~$0.0217 per standard page
Claude Opus 4.5	Anthropic	63.1%✓	5.1s	~59.2 pages / $1~$0.0169 per standard page
Mistral Medium	Mistral AI	62.1%✓	3.7s	~739 pages / $1~$0.00135 per standard page
Google: Gemini 3.1 Flash Lite	OOpenRouter	60.7%✓	1.5s	~784 pages / $1~$0.00128 per standard page
Gemini 3.1 Flash Lite Preview	Google Cloud	60.5%✓	1.1s	~845 pages / $1~$0.00118 per standard page
Qwen: Qwen3 VL 30B A3B Instruct	OOpenRouter	56.1%✓	4.1s	~1,940 pages / $1~$0.000517 per standard page
Qwen3-VL 30B A3B Instruct (DeepInfra)	DeepInfra	56.1%✓	4.5s	~2,470 pages / $1~$0.000404 per standard page
MiniMax: MiniMax M3	OOpenRouter	55%✓	4.8s	~940 pages / $1~$0.00106 per standard page
xAI: Grok 4.3	OOpenRouter	54.3%✓	4.7s	~318 pages / $1~$0.00314 per standard page
Anthropic: Claude Sonnet 4.6	OOpenRouter	54.1%✓	5.2s	~76.9 pages / $1~$0.013 per standard page
Meta: Llama 4 Scout	OOpenRouter	53.6%✓	1.6s	~1,370 pages / $1~$0.000731 per standard page
Mistral Small	Mistral AI	53.6%✓	1.6s	~4,560 pages / $1~$0.000219 per standard page
ByteDance Seed: Seed-2.0-Code	OOpenRouter	53%✓	15.4s	~259 pages / $1~$0.00385 per standard page
Qwen2.5-VL 32B Instruct (DeepInfra)	DeepInfra	51.6%✓	5.2s	~467 pages / $1$0.00214/page
Anthropic: Claude Sonnet 4.5	OOpenRouter	50.3%✓	6.3s	~76.9 pages / $1~$0.013 per standard page
Anthropic: Claude Sonnet 4	OOpenRouter	49.5%✓	3.8s	~76.9 pages / $1~$0.013 per standard page
Qwen: Qwen3.6 Flash	OOpenRouter	49%✓	15.4s	~1,270 pages / $1~$0.000789 per standard page
Anthropic: Claude Sonnet 5	OOpenRouter	47.4%✓	4.6s	~136 pages / $1~$0.00733 per standard page
Claude Opus 5	OOpenRouter	45.6%✓	5.0s	~45.2 pages / $1~$0.0221 per standard page
Qwen: Qwen3.6 27B	OOpenRouter	45.1%✓	10.4s	~365 pages / $1~$0.00274 per standard page
Qwen: Qwen3.6 35B A3B	OOpenRouter	43.8%✓	6.4s	~1,170 pages / $1~$0.000852 per standard page
OpenAI: GPT-4.1 Nano	OOpenRouter	43.5%✓	2.3s	~3,040 pages / $1~$0.000329 per standard page
Thinking Machines: Inkling Small	OOpenRouter	43.4%✓	2.5s	~389 pages / $1~$0.00257 per standard page
Anthropic: Claude Haiku 4.5	OOpenRouter	42.6%✓	3.4s	~274 pages / $1~$0.00365 per standard page
Claude Haiku 4.5	Anthropic	42.6%✓	2.5s	~300 pages / $1~$0.00333 per standard page
Amazon: Nova 2 Lite	OOpenRouter	41.6%✓	2.0s	~580 pages / $1~$0.00172 per standard page
SpaceXAI: Grok 4.6	OOpenRouter	41.1%✓	4.0s	~99.4 pages / $1~$0.0101 per standard page
Amazon: Nova Lite 1.0	OOpenRouter	40.1%✓	2.2s	~4,260 pages / $1~$0.000235 per standard page
GPT 4.1	OpenAI	36.1%✓	2.1s	~174 pages / $1~$0.00575 per standard page
GPT 4o Mini	OpenAI	32.3%✓	2.0s	~401 pages / $1~$0.00249 per standard page
OpenAI: GPT-5.4 Image 2	OOpenRouter	31.6%✓	2.5s	~48.2 pages / $1~$0.0207 per standard page
ByteDance: UI-TARS 7B	OOpenRouter	21.5%✓	3.5s	~4,180 pages / $1~$0.000239 per standard page
Amazon: Nova Pro 1.0	OOpenRouter	21.1%✓	2.4s	~320 pages / $1~$0.00313 per standard page
docTR	Open-OCR.com	12.8%✓	2.0s	500 pages / $1$0.002/page
EasyOCR	Open-OCR.com	4.8%✓	650ms	500 pages / $1$0.002/page
Anthropic Claude Haiku Latest	OOpenRouter	No hard score	4.6s	~231 pages / $1~$0.00433 per standard page
Anthropic Claude Sonnet Latest	OOpenRouter	No hard score	6.5s	~136 pages / $1~$0.00733 per standard page
Anthropic: Claude Fable 5.1	OOpenRouter	No hard score	—	~24.4 pages / $1~$0.041 per standard page
Anthropic: Claude Opus 4.8	OOpenRouter	No hard score	5.1s	~54.6 pages / $1~$0.0183 per standard page
Anthropic: Claude Sonnet 5.5	OOpenRouter	No hard score	—	~122 pages / $1~$0.00821 per standard page
ByteDance Seed: Seed 2.1 Turbo	OOpenRouter	No hard score	31.6s	~320 pages / $1~$0.00313 per standard page
Gemini 2.5 Flash-Lite	Google Cloud	No hard score	1.7s	~3,710 pages / $1~$0.00027 per standard page
Mistral Medium 3 5	Mistral AI	No hard score	1.6s	~166 pages / $1~$0.00603 per standard page
Mistral: Mistral Large 4	OOpenRouter	No hard score	—	~515 pages / $1~$0.00194 per standard page
Mistral: Mistral Medium 3.5	OOpenRouter	No hard score	4.0s	~182 pages / $1~$0.0055 per standard page
MoonshotAI: Kimi K2.6	OOpenRouter	No hard score	35.4s	~230 pages / $1~$0.00436 per standard page
Nex AGI: Nex-N2.5-Mini	OOpenRouter	No hard score	—	~11,600 pages / $1~$0.0000864 per standard page
OCRmyPDF	Open-OCR.com	No hard score	540ms	500 pages / $1$0.002/page
Pareto 26.10 Preview	OOpenRouter	No hard score	—	~362 pages / $1~$0.00276 per standard page
Qwen: Qwen3.5 Plus 2026-04-20	OOpenRouter	No hard score	14.0s	~415 pages / $1~$0.00241 per standard page
Qwen: Qwen3.7 Plus	OOpenRouter	No hard score	16.9s	~731 pages / $1~$0.00137 per standard page
RapidOCR	Open-OCR.com	No hard score	300ms	500 pages / $1$0.002/page
SpaceXAI: Grok 4.7	OOpenRouter	No hard score	—	~178 pages / $1~$0.00562 per standard page
Tesseract 5	Open-OCR.com	No hard score	420ms	500 pages / $1$0.002/page
Thinking Machines: Inkling	OOpenRouter	No hard score	3.5s	~240 pages / $1~$0.00416 per standard page
Z.ai: GLM Flash Latest	OOpenRouter	No hard score	—	~12,300 pages / $1~$0.0000815 per standard page
Xiaomi: MiMo-V2.5	OOpenRouter	57.3%✓	13.3s	~2,620 pages / $1~$0.000381 per standard page
xAI: Grok 4.5	OOpenRouter	56.5%✓	5.3s	~176 pages / $1~$0.00567 per standard page
Qwen: Qwen3.8 27B	OOpenRouter	47.3%✓	5.2s	~455 pages / $1~$0.0022 per standard page
xAI: Grok Build 0.1	OOpenRouter	45.1%✓	32.7s	~89.5 pages / $1~$0.0112 per standard page
OpenAI: GPT-5.1-Codex	OOpenRouter	33.5%✓	6.0s	~135 pages / $1~$0.0074 per standard page
Anthropic: Claude Opus 5.5	OOpenRouter	No hard score	—	~60.9 pages / $1~$0.0164 per standard page
Anthropic: Claude Opus Latest	OOpenRouter	No hard score	5.6s	~68.2 pages / $1~$0.0147 per standard page
MoonshotAI: Kimi K2.7 Code	OOpenRouter	No hard score	10.1s	~288 pages / $1~$0.00347 per standard page
Nex AGI: Nex-N2.5-Pro	OOpenRouter	No hard score	—	~4,410 pages / $1~$0.000227 per standard page
Sakana: Fugu Max	OOpenRouter	No hard score	—	~178 pages / $1~$0.00562 per standard page
Xiaomi: MiMo-V2.6-Pro-UltraSpeed	OOpenRouter	No hard score	—	~106 pages / $1~$0.0094 per standard page
Perplexity: Sonar Pro	OOpenRouter	37.1%✓	3.1s	~91.9 pages / $1~$0.0109 per standard page
Kimi K3 (Fireworks)	Fireworks AI	No hard score	—	~24.1 pages / $1~$0.0414 per standard page
MoonshotAI Kimi Latest	OOpenRouter	No hard score	46.2s	~68.5 pages / $1~$0.0146 per standard page
Qwen: Qwen3.8 Flash	OOpenRouter	No hard score	—	~2,200 pages / $1~$0.000454 per standard page
Qwen: Qwen3.8 Max (0902)	OOpenRouter	No hard score	—	~178 pages / $1~$0.00562 per standard page
Meta: Muse Spark 1.2	OOpenRouter	65.5%✓	7.6s	~103 pages / $1~$0.00975 per standard page
OpenAI: GPT-5.6 Luna	OOpenRouter	64.8%✓	4.8s	~816 pages / $1~$0.00123 per standard page
Mistral: Mistral Large 3 2512	OOpenRouter	58.5%✓	4.6s	~849 pages / $1~$0.00118 per standard page
Gemini 3.5 Flash Lite	Google Cloud	56.9%✓	1.3s	~544 pages / $1~$0.00184 per standard page
Google: Gemini 3.5 Flash-Lite	OOpenRouter	56.9%✓	1.7s	~504 pages / $1~$0.00198 per standard page
Gemini 3.6 Flash	Google Cloud	56.5%✓	4.1s	~179 pages / $1~$0.00559 per standard page
Google: Gemini 3.6 Flash	OOpenRouter	56.5%✓	4.4s	~189 pages / $1~$0.00529 per standard page
OpenAI: GPT-5.6 Luna Pro	OOpenRouter	53.9%✓	22.0s	~189 pages / $1~$0.0053 per standard page
Gemini 3.7 Flash	Google Cloud	51.6%✓	2.6s	~217 pages / $1~$0.0046 per standard page
Google: Gemini 3.7 Flash	OOpenRouter	51.6%✓	3.8s	~226 pages / $1~$0.00442 per standard page
Anthropic: Claude Fable 5	OOpenRouter	50.8%✓	6.8s	~27.3 pages / $1~$0.0367 per standard page
Meta: Muse Glimmer 30B	OOpenRouter	49.2%✓	7.6s	~754 pages / $1~$0.00133 per standard page
Gemini 3.5 Flash	Google Cloud	44.2%✓	4.2s	~51.3 pages / $1~$0.0195 per standard page
Google: Gemini 3.5 Flash	OOpenRouter	44.2%✓	5.9s	~59.6 pages / $1~$0.0168 per standard page
O4 Mini	OpenAI	41.7%✓	11.7s	~72.1 pages / $1~$0.0139 per standard page
Perceptron: Perceptron Mk1	OOpenRouter	41.2%✓	13.6s	~982 pages / $1~$0.00102 per standard page
OpenAI: GPT-5.6 Terra Pro	OOpenRouter	37.6%✓	14.2s	~50.7 pages / $1~$0.0197 per standard page
OpenAI: GPT-5.6 Sol	OOpenRouter	36.1%✓	6.4s	~139 pages / $1~$0.00721 per standard page
OpenAI: GPT-5.6 Sol Pro	OOpenRouter	33.7%✓	8.6s	~77.3 pages / $1~$0.0129 per standard page
Perplexity: Sonar	OOpenRouter	32.8%✓	2.5s	~966 pages / $1~$0.00104 per standard page
Gemini 3.1 Pro Preview	Google Cloud	30.4%✓	22.7s	~21.7 pages / $1~$0.046 per standard page
Gemini 3.1 Pro Preview Custom Tools	Google Cloud	30.4%✓	27.9s	~18.3 pages / $1~$0.0546 per standard page
o3	OpenAI	30.4%✓	5.8s	~54.7 pages / $1~$0.0183 per standard page
OpenAI: GPT-5.6 Terra	OOpenRouter	29.5%✓	2.6s	~118 pages / $1~$0.00851 per standard page
OpenAI: GPT-5.5	OOpenRouter	29.2%✓	7.6s	~35.6 pages / $1~$0.0281 per standard page
Gemini 2.5 Pro	Google Cloud	25.6%✓	27.0s	~25.1 pages / $1~$0.0398 per standard page
Google: Gemini 2.5 Pro Preview 06-05	OOpenRouter	25.6%✓	25.7s	~41.3 pages / $1~$0.0242 per standard page
Anthropic: Claude Fable Latest	OOpenRouter	No hard score	7.1s	~27.3 pages / $1~$0.0367 per standard page
Anthropic: Claude Opus 4.1	OOpenRouter	No hard score	15.4s	~18.3 pages / $1~$0.0547 per standard page
Fireworks: Ember-1	OOpenRouter	No hard score	—	~81.2 pages / $1~$0.0123 per standard page
Gemini 3.8 Flash	Google Cloud	No hard score	—	~351 pages / $1~$0.00285 per standard page
Google Gemini Flash Latest	OOpenRouter	No hard score	4.8s	~166 pages / $1~$0.00601 per standard page
Google Gemini Pro Latest	OOpenRouter	No hard score	18.3s	~27.6 pages / $1~$0.0362 per standard page
Google: Gemini 3.8 Flash	OOpenRouter	No hard score	—	~325 pages / $1~$0.00308 per standard page
Google: Nano Banana 2.1	OOpenRouter	No hard score	—	~162 pages / $1~$0.00616 per standard page
Meta: Muse Spark 1.1	OOpenRouter	No hard score	4.0s	~143 pages / $1~$0.007 per standard page
Meta: Muse Spark 1.3	OOpenRouter	No hard score	—	~261 pages / $1~$0.00383 per standard page
MoonshotAI: Kimi K3	OOpenRouter	No hard score	6.9s	~75.2 pages / $1~$0.0133 per standard page
o1	OpenAI	No hard score	12.8s	~10.8 pages / $1~$0.0922 per standard page
OpenAI GPT Astra Latest	OOpenRouter	No hard score	—	~24.4 pages / $1~$0.041 per standard page
OpenAI GPT Luna Latest	OOpenRouter	No hard score	—	~2,440 pages / $1~$0.00041 per standard page
OpenAI GPT Mini Latest	OOpenRouter	No hard score	1.8s	~313 pages / $1~$0.00319 per standard page
OpenAI GPT Sol Latest	OOpenRouter	No hard score	—	~122 pages / $1~$0.00821 per standard page
OpenAI GPT Terra Latest	OOpenRouter	No hard score	—	~105 pages / $1~$0.0095 per standard page
OpenAI: GPT Chat Latest	OOpenRouter	No hard score	3.1s	~47.6 pages / $1~$0.021 per standard page
OpenAI: GPT-5.5 Pro	OOpenRouter	No hard score	—	~7.88 pages / $1~$0.127 per standard page
OpenAI: GPT-6 Astra	OOpenRouter	No hard score	—	~24.4 pages / $1~$0.041 per standard page
OpenAI: GPT-6 Astra Pro	OOpenRouter	No hard score	—	~24.4 pages / $1~$0.041 per standard page
OpenAI: GPT-6 Sol	OOpenRouter	No hard score	—	~122 pages / $1~$0.00821 per standard page
OpenAI: GPT-6 Sol Pro	OOpenRouter	No hard score	—	~122 pages / $1~$0.00821 per standard page
OpenAI: GPT-6.1 Sol	OOpenRouter	No hard score	—	~122 pages / $1~$0.00821 per standard page
OpenAI: GPT-6.1 Sol Pro	OOpenRouter	No hard score	—	~122 pages / $1~$0.00821 per standard page
Qwen: Qwen3.8 Max	OOpenRouter	No hard score	14.9s	~157 pages / $1~$0.00636 per standard page
Qwen: Qwen3.8 Max Prime	OOpenRouter	No hard score	—	~89 pages / $1~$0.0112 per standard page
Sakana: Fugu Ultra	OOpenRouter	No hard score	43.1s	~13.4 pages / $1~$0.0748 per standard page
Sakana: Fugu Ultra v2	OOpenRouter	No hard score	—	~42.1 pages / $1~$0.0238 per standard page
Amazon: Nova Premier 1.0	OOpenRouter	No hard score	5.5s	~87 pages / $1~$0.0115 per standard page
Pareto Latest	OOpenRouter	No hard score	—	~142 pages / $1~$0.00702 per standard page
ScaleDown: Extract	OOpenRouter	No hard score	—	~28,900 pages / $1~$0.0000346 per standard page