Baidu Models

25 modelsGeneral models from $0.0136/M inputUp to 1M context

Usage

Last 29 days · 2026-10-22 to 2026-11-19

Tokens

702M

Requests

69.5K

Models in use

24 of 25

Tokens per day, stacked by model

0175M351M10-2210-2911-0511-1211-192026-10-22 — 1,140,620 tokens ernie-5.1: 1,003,400 qianfan-ocr: 119,735 3 more models: 7,780 ernie-5.0: 5,635 ernie-5.0-thinking-preview: 4,0702026-10-23 — 1,388,780 tokens ernie-5.1: 972,275 qianfan-ocr: 369,970 ernie-5.0: 38,695 ernie-5.0-thinking-preview: 3,980 3 more models: 3,8602026-10-24 — 1,234,225 tokens ernie-5.1: 925,540 ernie-5.0: 251,005 ernie-5.0-thinking-preview: 46,195 qianfan-ocr: 9,325 3 more models: 2,1602026-10-25 — 476,125 tokens ernie-5.1: 455,955 qianfan-ocr: 9,300 ernie-5.0: 5,265 ernie-5.0-thinking-preview: 3,355 3 more models: 2,2502026-10-26 — 826,650 tokens ernie-5.1: 663,345 qianfan-ocr: 149,240 ernie-5.0: 7,945 ernie-5.0-thinking-preview: 4,080 3 more models: 2,0402026-10-27 — 1,706,305 tokens ernie-5.0-thinking-preview: 1,142,050 ernie-5.1: 388,445 qianfan-ocr: 166,670 ernie-5.0: 6,425 3 more models: 2,7152026-10-28 — 49,063,500 tokens ernie-5.1: 42,328,945 ernie-5.0-thinking-preview: 6,651,825 qianfan-ocr: 74,860 ernie-5.0: 5,720 3 more models: 2,1502026-10-29 — 11,269,210 tokens ernie-5.1: 10,785,740 qianfan-ocr: 427,390 ernie-5.0: 51,155 ernie-5.0-thinking-preview: 3,215 3 more models: 1,7102026-10-30 — 28,679,275 tokens ernie-5.1: 28,581,290 qianfan-ocr: 97,9852026-10-31 — 14,340,790 tokens ernie-5.1: 12,769,315 musesteamer-air-image: 1,526,500 qianfan-ocr: 44,9752026-11-01 — 6,458,375 tokens ernie-5.1: 5,929,245 musesteamer-air-image: 443,750 3 more models: 85,3802026-11-02 — 1,400,370 tokens ernie-5.1: 957,700 qianfan-ocr: 428,355 3 more models: 14,3152026-11-03 — 1,910,845 tokens ernie-5.1: 839,080 pp-structurev3: 500,000 qianfan-ocr: 391,335 3 more models: 180,4302026-11-04 — 6,350,675 tokens ernie-5.1: 6,350,6752026-11-05 — 7,967,720 tokens ernie-5.1: 7,868,445 musesteamer-air-image: 71,000 qianfan-ocr: 28,265 3 more models: 102026-11-06 — 15,898,730 tokens ernie-5.1: 14,937,130 qianfan-ocr: 517,850 musesteamer-air-image: 443,7502026-11-07 — 1,681,610 tokens musesteamer-air-image: 816,500 ernie-5.1: 813,000 qianfan-ocr: 52,070 3 more models: 402026-11-08 — 1,052,875 tokens ernie-5.1: 1,030,560 qianfan-ocr: 22,275 3 more models: 402026-11-09 — 1,413,090 tokens ernie-5.1: 1,241,755 qianfan-ocr: 171,3352026-11-10 — 6,930,850 tokens ernie-5.1: 6,522,520 musesteamer-air-image: 372,750 qianfan-ocr: 35,5802026-11-11 — 15,256,415 tokens ernie-5.1: 15,225,695 qianfan-ocr: 29,975 ernie-5.5-flash-preview: 7452026-11-12 — 350,801,890 tokens ernie-5.5-flash-preview: 346,529,575 ernie-5.5-flash-preview-free: 4,110,630 ernie-5.1: 161,6852026-11-13 — 18,584,215 tokens ernie-5.5-flash-preview-free: 17,799,095 ernie-5.1: 473,590 ernie-5.5-flash-preview: 212,650 qianfan-ocr: 98,8802026-11-14 — 21,679,750 tokens ernie-5.5-flash-preview-free: 20,663,885 ernie-5.1: 867,320 musesteamer-air-image: 71,000 qianfan-ocr: 56,720 ernie-5.5-flash-preview: 20,8252026-11-15 — 15,117,260 tokens ernie-5.5-flash-preview-free: 12,459,435 ernie-5.1: 1,617,290 ernie-5.5-flash-preview: 946,165 qianfan-ocr: 94,350 3 more models: 202026-11-16 — 43,188,080 tokens ernie-5.5-flash-preview-free: 22,443,560 ernie-5.5-flash-preview: 20,298,925 ernie-5.1: 445,5952026-11-17 — 57,495,015 tokens ernie-5.5-flash-preview: 46,236,895 ernie-5.5-flash-preview-free: 10,909,685 ernie-5.1: 330,950 qianfan-ocr: 17,4852026-11-18 — 11,241,610 tokens ernie-5.1: 6,177,080 ernie-5.5-flash-preview-free: 4,221,735 ernie-5.5-flash-preview: 810,720 qianfan-ocr: 32,0752026-11-19 — 7,944,970 tokens ernie-5.5-flash-preview-free: 4,999,680 ernie-5.1: 2,753,715 ernie-5.5-flash-preview: 151,990 qianfan-ocr: 39,585
  • ernie-5.5-flash-preview
  • ernie-5.1
  • ernie-5.5-flash-preview-free
  • ernie-5.0-thinking-preview
  • musesteamer-air-image
  • qianfan-ocr
  • pp-structurev3
  • ernie-5.0
  • 3 more models

Which models that traffic went to

  1. ERNIE 5.5 Flash Preview59.1%415M
  2. ERNIE 5.124.7%173M
  3. ERNIE 5.5 Flash Preview (free)13.9%97.6M
  4. ERNIE 5.0 Thinking Preview1.1%7.9M
  5. Musesteamer Air Image0.5%3.7M
  6. Qianfan Ocr0.5%3.5M
  7. Pp Structurev30.1%500K
  8. ERNIE 5.00.1%372K
  9. 3 more models<0.1%305K

Share of 702M tokens. 13 models with traffic report no token counts and cannot be ranked here, including ernie-4.5-0.3b and ernie-4.5-turbo-latest — they are in the request view.

The two views disagree on purpose: a model can take a large share of the calls and a small share of the tokens — many short requests — or the reverse. Which one matters depends on whether your cost is driven by call volume or by prompt length. Measured on AIHubMix over the last 29 days, counting the 25 model IDs listed on this page; traffic routed through upstream-specific IDs that are not in the public catalog is not included.

All 25 Baidu Models

Open in model list
Baidu models on AIHubMix with input and output modalities, context length, maximum output, price per million tokens including cache read and cache write rates, and measured throughput and latency.
Modalities
ernie-5.5-flash-previewTakes text, returns text.1M—$0.294$1.176/M$0.0294/M62 tok/s4.75 s
ernie-5.0-thinking-previewTakes text, returns text.183K64K$0.822$3.288/M$0.822/M6 tok/s2.20 s
ernie-4.5Takes text, vision, returns text.160K64K$0.068$0.272/M—99 tok/s0.95 s
ernie-4.5-turbo-vlTakes text, vision, returns text.139K16K$0.4$1.2/M—9 tok/s1.29 s
ernie-4.5-turbo-latestTakes text, vision, returns text.135K12K$0.11$0.44/M—18 tok/s0.62 s
ERNIE-X1.1-PreviewTakes text, returns text.119K64K$0.136$0.544/M—4 tok/s1.26 s
ernie-5.1Takes text, returns text.119K66K$0.5634$2.5353/M$0.5634/M6 tok/s2.83 s
ernie-5.0Takes text, vision, returns text.119K66K$0.8219$3.2877/M$0.8219/M5 tok/s2.29 s
ernie-5.0-thinking-expTakes text, returns text.119K66K$0.8219$3.2877/M$0.8219/M——
ernie-x1-turboTakes text, returns text.51K28K$0.136$0.544/M———
qianfan-ocrTakes text, vision, returns text.32K28K$0.062$0.248/M—20 tok/s0.38 s
qianfan-ocr-fastTakes text, vision, returns text.32K28K$0.664$2.7383/M———
ernie-5.5-flash-preview-freeTakes text, returns text.——FreeFree/M———
ernie-4.5-0.3bTakes text, vision, returns text.——$0.0136$0.0544/M———
embedding-v1Takes text. Output modality not published.——$0.068$0.068/M———
tao-8k——$0.068$0.068/M———
ernie-4.5-turbo-128k-previewTakes text, vision, returns text.——$0.108$0.432/M———
ernie-x1.1-previewTakes , returns text.——$0.136$0.544/M—4 tok/s1.26 s
qianfan-qi-vlTakes , returns text.——$0.2$0.6/M———
baidu/ERNIE-4.5-300B-A47BTakes text, vision, returns text.——$0.32$1.28/M———
cc-ernie-4.5-300b-a47bTakes text, vision, returns text.——$0.32$1.28/M———
ernie-image-turboTakes , returns vision.——$0.0155 / img———
musesteamer-air-imageTakes , returns vision.——$0.0071 / img———
paddleocr-vl-0.9bTakes , returns text.——$0.025 / page———
pp-structurev3Takes , returns text.——$0.025 / page———

Prices are USD per million tokens; cache read and cache write are the rates for prompt-cache hits and for writing a prompt into the cache. Throughput and latency are measured on AIHubMix — the same figures the model detail page shows — not vendor claims. A dash means the catalog does not publish that field for that model, which is not the same as the model not supporting it.

Baidu on AIHubMix

Which Baidu model should I start with?

ernie-4.5-0.3b at $0.0136/M input — the cheapest entry here that declares tool calling. Move up to ernie-5.0-thinking-preview when answer quality matters more than cost.

Which of these models reason before answering?

6 of the 25 models here declare a reasoning phase — they work through the problem before producing an answer, which helps on multi-step problems at the cost of extra output tokens. Use the Reasoning filter above the table to see them. The catalog does not record anything further about how they differ, so this page does not sort them into families.

Why are there several entries for the same model?

Because each row is a route you can call, not a model release. Some IDs name an upstream (azure-, alicloud-, cc-), some are the open-weight repository form (baidu/…), and some differ only in capitalisation, kept so older integrations keep working.

The catalog does not carry a field saying which of those a given row is, so this page does not sort them into buckets it would have to invent. Every row shows that route’s own price, context and speed — compare those directly, and open a model to see the upstreams that serve it.

How is cached input billed?

The Cache read column is the rate for input tokens served from the prompt cache — for example ernie-5.5-flash-preview bills cache hits at 10% of the input rate. Cache write is the surcharge for putting a prompt into the cache in the first place, and only a few upstreams bill it separately. A dash in either column means the catalog carries no cache rate for that model, so plan on paying the full input rate.

Do I need a separate Baidu account?

No. One AIHubMix key covers every model on this page, and switching between them is a change to the model string — billing, rate limits, and logs stay in one place.

Start calling Baidu in one line

One key, one endpoint, 984 models across 49 model authors.