LLMbaidu

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 preserving versatile multimodal intelligence, it provides a powerful performance upgrade over Qianfan-OCR.

Anyone in the Space can @-mention Baidu: Qianfan-OCR-Fast with the team's shared context — pooled credits, one chat, one memory.

All models

Starter is free forever — 1 Space, 100 credits/month, 1 MCP. No card.

Verdict

Qianfan-OCR-Fast is Baidu's specialized model for extracting text from images at speed. With a 65K context window and aggressive pricing ($0.68/$2.81 per Mtok), it handles document processing workflows where OCR accuracy matters more than general reasoning. This is not a conversational model — reach for it when you need structured text extraction from receipts, forms, screenshots, or scanned documents at scale.

Best for

High-volume receipt and invoice processing
Extracting text from Chinese-language documents
Form data extraction with structured output
Screenshot text capture for automation
Cost-sensitive document digitization pipelines

Strengths

Pricing undercuts most vision-capable models by 40-60% on input tokens, making it viable for batch document jobs. The 65K window accommodates multi-page PDFs or long-form scans in a single call. Baidu's OCR heritage means strong performance on mixed Chinese-English text and complex layouts like tables or forms. Fast inference lives up to the name — suitable for real-time extraction in user-facing apps.

Trade-offs

No public benchmarks make it hard to compare accuracy against Google Document AI or Azure Vision. This model does OCR, not reasoning — don't expect it to answer questions about document content or perform analysis beyond text extraction. Baidu's ecosystem means fewer integrations outside China-focused stacks. The proprietary license limits transparency into training data and fine-tuning options.

Specifications

Provider: baidu
Category: llm
Context length: 65,536 tokens
Max output: 28,672 tokens
Modalities: image, text
License: proprietary
Released: 2026-04-20

Pricing

Input: $0.68/Mtok
Output: $2.81/Mtok
Model ID: baidu/qianfan-ocr-fast

Per-token prices show what the model costs upstream. On Switchy your team draws from one shared org credit pool — one plan, one balance for everyone.

Team cost calculator

Seats5 peopleMessages / seat / day80Avg turn size2 ktokOutput share30 %

Estimated monthly spend

$23.21

17.6M tokens / month
5 seats · 80 msgs/day

Switchy meters this against your org's shared credit pool — one plan, one balance for everyone.

Providers

Provider-level routing data is not available yet for this model.

Performance

Performance snapshots are collected daily. Check back after the next ingestion run.

Benchmarks

Public benchmark scores are not available yet for this model. Check back after the next ingestion run.

Works well with

Top MCPs

Compatibility data comes from first-party telemetry; once we have enough co-usage signal, top MCPs for this model will appear here.

How Switchy teams use it

Not enough Spaces have used this model yet to share anonymised team stats. We wait for at least 50 distinct Spaces per week before publishing any aggregate.

Starter prompts

Extract Invoice Line Items

Extract all line items from this invoice image. Return vendor name, invoice date, total amount, and a table of items with descriptions, quantities, and prices.

Open in a Space →

Digitize Handwritten Forms

Read the handwritten text from this form and output each field as a JSON object with field names as keys and handwritten values as strings.

Open in a Space →

Screenshot Text Capture

Extract all text visible in this screenshot. Preserve the reading order from top to bottom, left to right, and note any button labels or menu items.

Open in a Space →

Table Data from Scans

Extract the table from this scanned document and format it as CSV. Preserve all column headers and ensure each row aligns correctly.

Open in a Space →

Multi-Language Receipt OCR

Extract merchant name, date, items purchased, and total amount from this receipt. Handle both Chinese and English text, and return amounts in their original currency.

Open in a Space →