Intelligent document processing for small business buyers
Asked as “intelligent document processing platform”, and as “IDP software”, on behalf of a small B2B company. 50 first choices recorded across the direct, paraphrase, budget and scale prompts, twelve models each. Added to the September 2026 Edition; its answers are read by the distilled judge (ai-indexes-judge-qwen3-14b-run3), not the claude-opus-5 judge of the earlier categories: how the two compare.
Standing · first-choice share
22%
Contested · Nanonets 16%
22Docsumo16Nanonets12Parseur50others
22% of first choices, contested.
By buyer segment
The same question asked on behalf of a different buyer. Each standing is computed within its segment; they sit side by side and are never added together.
The standing
Share is the count of first choices across the direct, paraphrase, budget and scale prompts, over all twelve models, for a small B2B company. Ordered by share.
ProductFirst-choice shareNegative rateLabelsQuadrant
Show the four products at 0%, ordered by negative rate
Bars are the share of first choices, 0 to 100Every product with at least 10 labels here. Every product name links to its product page.
Google Document AI is made by Google, whose model Gemini 3.5 Flash is in the set. On the four share prompts that model made it the first choice zero times of 4; the other eleven models three times of 44. Lab treatment is defined on the method page; the row is marked, not excluded.
Recommended versus criticized
Every product with at least 10 labels here, on both axes. The 30% line names a quadrant, not the verdict above: that one needs more than 40%.
Criticized challengerCriticized default
Negative label rate →
01
02
03G04
05
06
07A08
A09
10
11
12
13
14Accepted challengerEndorsed leader
0%First-choice share → · lines at 30% share and 25% negative40%
Key
01Docsumo22%
Docsumo22%02Nanonets16%
Nanonets16%03Parseur12%
Parseur12%04Google Document AI8%
05Lido8%
Lido8%06Docparser4%
Docparser4%07Rossum4%
Rossum4%08Amazon Textract2%
09Azure AI Document Intelligence2%
10ABBYY Vantage2%
ABBYY Vantage2%11Klippa DocHorizon0%
Klippa DocHorizon0%12Automation Anywhere0%
Automation Anywhere0%13Hyperscience Hypercell0%
Hyperscience Hypercell0%14UiPath Document Understanding0%
UiPath Document Understanding0%What they warned about
Two of twelve models held their first choice under the paraphrase. GPT-5.4 mini, Gemini 3.5 Flash, Grok 4.1 Fast, Mistral Small, DeepSeek V4 Flash, Llama 4 Maverick, Qwen 3.7 Flash, Kimi K2, GLM 4.7 FlashX and MiniMax M2.5 changed. A high negative share on a product with few labels is a warning. A low share on a product with many labels is salience, not sentiment.
ABBYY Vantage68%
21 of 31 labels negative · 10 of 12 models · 11 hard negative
“What to avoid on a tight budget ... Enterprise AI-driven IDP platforms like Rossum, ABBYY Vantage, Hyperscience, etc., are quote-based” Muse Glimmer 30B, budget prompt
UiPath Document Understanding62%
16 of 26 labels negative · 10 of 12 models · 8 hard negative
“The biggest cautionary candidates for small businesses are the enterprise-tier platforms — **UiPath, ABBYY, Automation Anywhere, Hyperscience, and Tungsten**” DeepSeek V4 Flash, negative prompt
Rossum39%
16 of 41 labels negative · 10 of 12 models · 7 hard negative
“What to Avoid (For Now) ... **Rossum** | Starts at $18,000/year—designed for mid-market/enterprise” Kimi K2, paraphrase prompt
Automation Anywhere90%
9 of 10 labels negative · 8 of 12 models · 4 hard negative
“The pricing model is described as **expensive and complex**, making it difficult for small teams to budget” Mistral Small, negative prompt
What they cite
Citations exist only for the models that return a source list: fourteen of the twelve in this edition, and all six flagship models on the expanded tier.
Sites the answers cite
76 of 84 answers in this category came back with a source list, from 14 of 12 models: citations where the model returns them, or the search results it consulted. 1063 links across 217 sites, every framing counted. Ranked by the number of answers carrying the site or page.
vendor site · G244 answers · 72 citations · 13 models
vendor site · Docsumo38 answers · 81 citations · 13 models
vendor site · Klippa34 answers · 40 citations · 11 models
vendor site · LarkSuite33 answers · 33 citations · 11 models
vendor site · Nanonets31 answers · 40 citations · 10 models
29 answers · 42 citations · 11 models
vendor site · FileCenter24 answers · 26 citations · 9 models
vendor site · Parseur22 answers · 39 citations · 7 models
vendor site · Lido21 answers · 27 citations · 10 models
vendor site · RaftLabs21 answers · 21 citations · 9 models
vendor site · Flowwright17 answers · 19 citations · 7 models
vendor site · FitGap16 answers · 22 citations · 10 models
Pages the answers cite
The ten pages named in the most answers, by full address. A page here is one the models returned with a recommendation, not one the index endorses.
33 answers · 11 models
32 answers · 11 models
29 answers · 10 models
29 answers · 11 models
22 answers · 10 models
21 answers · 8 models
21 answers · 8 models
21 answers · 9 models
15 answers · 7 models
13 answers · 6 models
Search against answers
Each company's standing in the answers beside its site's footprint in Google search, one row a site: the products the models named on it with their shares, and the share they add up to; monthly searches on Google, and DataForSEO's estimate of AI search demand (modeled from search signals, directional, not a count of queries to any assistant), for the most-searched of the company's and its products' names (the name is in each row's hover text); estimated monthly organic visits to the site; and its best position in Google's top ten for “intelligent document processing platform”, “intelligent document processing platforms”. US estimates from DataForSEO and Google's Ads Transparency Center. A small company's site, or a mid-sized company's site for its flagship, is marked company; a product on a large parent's site (Google, Microsoft) has no site figures. A column with no figures for this category is left out, and an empty cell means none were seen, not none exist. Two measurements side by side: neither is read as the cause of the other.
| Company and products | Share | Own site cited | Name searches, Google | AI search demand, est. | Organic visits | Google ads, last 30 days |
|---|---|---|---|---|---|---|
| Docsumo docsumo.com | 22% | 73 | 590 | 3 | 12,362 | 0 |
| Nanonets nanonets.com | 16% | 84 | 1,900 | 29 | 7,984 | 2 |
| Parseur parseur.com | 12% | 29 | 880 | 9 | 17,087 | |
| Google Cloud parent site Google Document AI 8% | 8% | 72 | 165,000 | 7,596 | ||
| Lido lido.app | 8% | 14 | 33,100 | 9,264 | 1,134 | 1 |
| Microsoft parent site Microsoft Entra ID 2%, Microsoft Syntex 2%, Microsoft Azure Document Intelligence 2% | 6% | 480 | 2,240,000 | 403,136 | ||
| Docparser docparser.com | 4% | 23 | 880 | 6 | 3,199 | 1 |
| WorkOS workos.com | 4% | 99 | 12,100 | 107 | 33,478 | 75 |
| Rossum rossum.ai | 4% | 101 | 1,600 | 2,838 | 3,120 | 24 |
| Amazon parent site Amazon Textract 2% | 2% | 74 | 151,000,000 | 1,380,939 |
Google's top ten for the two searches hold 17 sites; 1 of them are among the sites the models cited here (g2.com). The first result for “intelligent document processing platform” is extend.ai.
How they answered
Six framings of the same buying question, each sent to every model in a fresh session with search on. One row per model, so a row shows whether it held its answer under rewording, what it named when cost was the constraint, and what it argued against. Computed from the raw judge labels.ShowHide
How they answered
Six framings of the same buying question, each sent to every model in a fresh session with search on. One row per model, so a row shows whether it held its answer under rewording, what it named when cost was the constraint, and what it argued against. Computed from the raw judge labels.
ShowHideModelDirect“What is the best intelligent document processing platform for a small B2B company?”Paraphrase“Which IDP software would you recommend to a small business that sells to other businesses?”Comparative“What are the top intelligent document processing platforms for a small team and how do they differ?”Budget-constrained“What is the best intelligent document processing platform for a small company with a tight budget?”Scale-constrained“We are a 30 person company evaluating an intelligent document processing platform. What should we look at?”Negative“Which intelligent document processing platforms should a small business avoid or be cautious about?”
Claude Haiku 4.5DocparserDocparserHeldDocsumo, NanonetsDocsumo, Google Document AIno first choiceagainst: Coupa, Docsumo, Nanonets, Rossum, Tipalti, Tungsten Automation
Five alternatives
Amazon Textract, Docsumo, Doxis AI.dp, Doxis DocHorizon, RossumThree alternatives
Doxis DocHorizon, Nanonets, RossumSix alternatives
Amazon Textract, Azure AI Document Intelligence, Docparser, Doxis, Google Document AI, RossumTwo alternatives
ABBYY FlexiCapture, DocparserGPT-5.4 miniABBYY VantageMicrosoft Azure Document IntelligenceChangedMicrosoft Syntexagainst: Azure Content UnderstandingGoogle Document AIagainst: ABBYY Vantage, Rossumno first choiceagainst: OpenText Intelligent Capture, Thomson Reuters Document Intelligence, UiPath Agentic Automation / UiPath Platform
Three alternatives
BILL AP/AR, Dext, Google Document AITwo alternatives
ABBYY Vantage, Google Document AIThree alternatives
Amazon Textract, Azure AI Document Intelligence, Google Document AITwo alternatives
Docparser, Microsoft document processing / SyntexGemini 3.5 FlashNanonetsagainst: ABBYY Vantage, Hyperscience Hypercell, RossumPropelAuthChangedagainst: Auth0Nanonetsagainst: ABBYY Vantage, Hyperscience Hypercell, UiPath Document UnderstandingAirparseragainst: ABBYY Vantage, Amazon Textract, Hyperscience Hypercellno first choiceagainst: ABBYY Vantage, Kofaxagainst: ABBYY FlexiCapture / Vantage, Amazon Textract, Automation Anywhere IQ Bot, Google Document AI, Hyperscience Hypercell, Microsoft Azure AI Document Intelligence, Tungsten Automation (formerly Kofax) TotalAgility, UiPath Document Understanding
Four alternatives
Airparser, Docsumo, Lido, ParsioFive alternatives
Google Workspace, JumpCloud, Kinde, Microsoft Entra ID, WorkOSFour alternatives
Affinda, Docparser, Docsumo, LidoFive alternatives
ExtractBee, Google Document AI, ImageToTable.ai, Lido, Paperless-ngxPerplexity SonarDocsumoDocsumoHeldLido, Parseuragainst: ABBYY Vantage, Automation Anywhere, UiPath Document UnderstandingParseuragainst: ABBYY Vantage, Nanonetsno first choiceagainst: Azure AI Document Intelligence, Workato
Five alternatives
ABBYY Vantage, Nanonets, Parseur, Rossum, UiPath Document UnderstandingFour alternatives
ABBYY Vantage, Parseur, Rossum, UiPath Document UnderstandingThree alternatives
Docsumo, Nanonets, RossumTwo alternatives
Docsumo, Microsoft AI BuilderGrok 4.1 FastParseuragainst: Amazon Textract, Azure AI Document Intelligence, Google Document AIDocsumoChangedDocparser, Parseuragainst: ABBYY Vantage, Klippa DocHorizon, Nanonets, Rossum, UiPath Document UnderstandingParseuragainst: ABBYY Vantage, Hypatos, Rossumno first choiceagainst: ABBYY Vantage, Automation Anywhere, Hyperscience Hypercell, UiPath Document Understanding
Three alternatives
Docparser, Docsumo, NanonetsFour alternatives
Mindee, Nanonets, Parseur, RossumOne alternative
DocsumoThree alternatives
Azure AI Document Intelligence, Google Document AI, NanonetsMistral SmallDocuWareWorkOSChangedABBYY Vantage, DocsumoLidoagainst: UiPath Document Understandingno first choiceagainst: ABBYY Vantage, Automation Anywhere, Kofax, Nanonets, UiPath Agentic Automation
Two alternatives
Docsumo, Klippa DocHorizonThree alternatives
Auth0, Microsoft Entra ID, OktaFour alternatives
Amazon Textract, Azure AI Document Intelligence, LlamaParse, UnstractDeepSeek V4 FlashDocsumoagainst: RossumNanonetsChangedagainst: RossumParseuragainst: RossumParseuragainst: ABBYY Vantage, Amazon Textract, Azure AI Document Intelligence, Google Document AI, Rossum, UiPath Document UnderstandingLido, Nanonets, Rossumagainst: ABBYY Vantage, Hyperscience Hypercell, Tungsten, UiPath Document Understandingagainst: ABBYY Vantage, Automation Anywhere, Hyperscience Hypercell, Tungsten Automation, UiPath Document Understanding
One alternative
ParseurTwo alternatives
Docsumo, Klippa DocHorizonThree alternatives
Docparser, Docsumo, NanonetsOne alternative
DocuClipperTwo alternatives
Docsumo, Klippa DocHorizonLlama 4 Maverickno first choiceDocsumoChangedno first choiceno first choiceno first choicenothing named
Two alternatives
Docparser, ParseurQwen 3.7 FlashNanonetsagainst: Automation Anywhere, UiPath Document UnderstandingBill.comChangedagainst: Automation Anywhere, UiPath Document UnderstandingNanonetsagainst: Amazon Textract, Azure AI Document Intelligence, Google Document AINanonetsno first choiceagainst: ABBYY FineReader Server, Amazon Textract, Azure AI Document Intelligence, Google Document AI, Kofax, Make, NICE, OCR.space, Salesforce, Tesseract, Zapier
Two alternatives
Docsumo, RossumTwo alternatives
Dext, KlearStackTwo alternatives
Docsumo, RossumTwo alternatives
DocuClipper, Google Document AIKimi K2Docsumo, Larkagainst: ABBYY Vantage, Hyperscience Hypercell, UiPath Document UnderstandingLidoChangedagainst: ABBYY Vantage, Hyperscience Hypercell, RossumDocsumoAmazon Textract, Azure AI Document Intelligenceno first choiceagainst: ABBYY FlexiCapture / Vantage, Automation Anywhere, UiPath Document Understanding
Two alternatives
FileCenter, RossumTwo alternatives
Docsumo, Klippa DocHorizonFive alternatives
Amazon Textract, Docparser, Microsoft Azure AI Document Intelligence, Nanonets, ParseurTwo alternatives
Docparser, NanonetsGLM 4.7 FlashXLidoMicrosoft Entra IDChangedDocsumo, NanonetsParseuragainst: ABBYY Vantage, Rossumno first choiceagainst: ABBYY FlexiCapture, Automation Anywhere, Hyland IDP, Indico, PDF.ai, UiPath Document Understanding
Two alternatives
Docsumo, ParseurOne alternative
OktaFour alternatives
ABBYY Vantage, Hyperscience Hypercell, Microsoft Syntex, RossumTwo alternatives
Docsumo, NanonetsMiniMax M2.5Docsumo, Nanonets, RossumDocsumo, NanonetsChangedLido, Rossumagainst: ABBYY Vantage, Hyperscience HypercellGoogle Document AI, Microsoft Syntexno first choiceagainst: ABBYY Vantage, Automation Anywhere, Rossum, UiPath Document Understanding
One alternative
Iron Mountain InSight® DXPOne alternative
DocuWareThree alternatives
Google Document AI, Microsoft Azure AI Document Intelligence, NanonetsTwo alternatives
Rossum, UiPath Document UnderstandingBold is the first choiceAlternatives are counted; the count opens them.What the answer argued against
The record
One row per call: the version string exactly as returned, whether the model searched, sources cited, and latency. Full answer text is in the free responses file. Download the record
Seventy-two rows: every prompt, every model, every answer.
| Prompt | Model | Version string | Time (UTC) | Searched | Sources | Latency |
|---|---|---|---|---|---|---|
| Direct recommendation | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-10-05 01:29 | yes | 9 | 8 s |
| Direct recommendation | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-10-05 01:02 | yes | 4 | 5 s |
| Direct recommendation | Gemini 3.5 Flash | gemini-3.5-flash | 2026-10-05 01:05 | yes | 14 | 21 s |
| Direct recommendation | Perplexity Sonar | sonar | 2026-10-04 23:49 | yes | 20 | 3 s |
| Direct recommendation | Grok 4.1 Fast | spacexai/grok-4.1-fast-non-reasoning via vertex | 2026-10-05 00:24 | yes | 19 | 10 s |
| Direct recommendation | Mistral Small | mistral/mistral-small via mistral | 2026-10-05 00:13 | yes | 5 | 4 s |
| Direct recommendation | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-10-05 00:48 | yes | 21 | 37 s |
| Direct recommendation | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-10-04 22:10 | yes | 5 | 2 s |
| Direct recommendation | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-10-05 00:28 | yes | 9 | 27 s |
| Direct recommendation | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-10-05 00:54 | yes | 14 | 29 s |
| Direct recommendation | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-10-04 22:39 | yes | 23 | 57 s |
| Direct recommendation | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-10-04 22:13 | yes | 5 | 17 s |
| Paraphrase | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-10-04 23:03 | yes | 9 | 8 s |
| Paraphrase | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-10-05 00:42 | yes | 3 | 4 s |
| Paraphrase | Gemini 3.5 Flash | gemini-3.5-flash | 2026-10-05 01:56 | yes | 7 | 25 s |
| Paraphrase | Perplexity Sonar | sonar | 2026-10-04 23:13 | yes | 16 | 3 s |
| Paraphrase | Grok 4.1 Fast | spacexai/grok-4.1-fast-non-reasoning via vertex | 2026-10-05 01:11 | yes | 22 | 8 s |
| Paraphrase | Mistral Small | mistral/mistral-small via mistral | 2026-10-04 22:13 | yes | 5 | 5 s |
| Paraphrase | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-10-05 00:46 | yes | 23 | 31 s |
| Paraphrase | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-10-05 01:06 | yes | 5 | 2 s |
| Paraphrase | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-10-05 00:58 | no | 0 | 26 s |
| Paraphrase | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-10-04 23:07 | yes | 24 | 19 s |
| Paraphrase | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-10-05 00:44 | no | 0 | 9 s |
| Paraphrase | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-10-05 00:02 | yes | 8 | 34 s |
| Comparative | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-10-05 01:15 | yes | 19 | 13 s |
| Comparative | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-10-05 00:33 | yes | 4 | 8 s |
| Comparative | Gemini 3.5 Flash | gemini-3.5-flash | 2026-10-05 00:39 | yes | 23 | 28 s |
| Comparative | Perplexity Sonar | sonar | 2026-10-04 22:55 | yes | 19 | 4 s |
| Comparative | Grok 4.1 Fast | spacexai/grok-4.1-fast-non-reasoning via vertex | 2026-10-05 01:26 | yes | 23 | 9 s |
| Comparative | Mistral Small | mistral/mistral-small via mistral | 2026-10-04 22:34 | yes | 13 | 11 s |
| Comparative | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-10-04 23:39 | yes | 23 | 49 s |
| Comparative | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-10-05 01:18 | yes | 5 | 2 s |
| Comparative | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-10-05 01:49 | yes | 9 | 40 s |
| Comparative | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-10-05 00:41 | yes | 22 | 23 s |
| Comparative | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-10-05 00:17 | yes | 22 | 36 s |
| Comparative | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-10-05 01:06 | yes | 5 | 21 s |
| Budget constrained | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-10-04 23:15 | yes | 9 | 8 s |
| Budget constrained | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-10-05 01:15 | yes | 4 | 6 s |
| Budget constrained | Gemini 3.5 Flash | gemini-3.5-flash | 2026-10-05 00:37 | yes | 16 | 29 s |
| Budget constrained | Perplexity Sonar | sonar | 2026-10-05 00:58 | yes | 19 | 3 s |
| Budget constrained | Grok 4.1 Fast | spacexai/grok-4.1-fast-non-reasoning via vertex | 2026-10-05 00:33 | yes | 25 | 13 s |
| Budget constrained | Mistral Small | mistral/mistral-small via mistral | 2026-10-05 00:14 | yes | 5 | 3 s |
| Budget constrained | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-10-04 22:53 | yes | 23 | 24 s |
| Budget constrained | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-10-04 23:54 | yes | 5 | 2 s |
| Budget constrained | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-10-05 00:59 | yes | 18 | 84 s |
| Budget constrained | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-10-04 23:56 | yes | 23 | 32 s |
| Budget constrained | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-10-05 00:44 | yes | 23 | 202 s |
| Budget constrained | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-10-05 01:46 | yes | 5 | 17 s |
| Scale constrained | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-10-04 23:59 | no | 0 | 6 s |
| Scale constrained | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-10-05 00:44 | no | 0 | 8 s |
| Scale constrained | Gemini 3.5 Flash | gemini-3.5-flash | 2026-10-05 00:32 | no | 0 | 15 s |
| Scale constrained | Perplexity Sonar | sonar | 2026-10-04 23:04 | yes | 22 | 4 s |
| Scale constrained | Grok 4.1 Fast | spacexai/grok-4.1-fast-non-reasoning via vertex | 2026-10-05 00:17 | yes | 15 | 9 s |
| Scale constrained | Mistral Small | mistral/mistral-small via mistral | 2026-10-04 23:39 | no | 0 | 7 s |
| Scale constrained | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-10-05 00:19 | yes | 25 | 40 s |
| Scale constrained | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-10-04 23:19 | yes | 5 | 3 s |
| Scale constrained | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-10-04 23:55 | no | 0 | 24 s |
| Scale constrained | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-10-05 01:11 | yes | 18 | 24 s |
| Scale constrained | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-10-04 22:58 | yes | 15 | 97 s |
| Scale constrained | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-10-05 01:47 | yes | 10 | 16 s |
| Negative framing | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-10-04 23:15 | yes | 18 | 8 s |
| Negative framing | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-10-04 23:09 | yes | 3 | 4 s |
| Negative framing | Gemini 3.5 Flash | gemini-3.5-flash | 2026-10-05 01:23 | yes | 16 | 24 s |
| Negative framing | Perplexity Sonar | sonar | 2026-10-05 01:26 | yes | 17 | 4 s |
| Negative framing | Grok 4.1 Fast | spacexai/grok-4.1-fast-non-reasoning via vertex | 2026-10-05 00:59 | yes | 24 | 11 s |
| Negative framing | Mistral Small | mistral/mistral-small via mistral | 2026-10-05 00:43 | yes | 13 | 8 s |
| Negative framing | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-10-05 01:45 | yes | 23 | 54 s |
| Negative framing | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-10-04 23:42 | yes | 5 | 3 s |
| Negative framing | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-10-04 23:52 | no | 0 | 27 s |
| Negative framing | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-10-04 22:50 | yes | 24 | 22 s |
| Negative framing | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-10-05 00:28 | yes | 23 | 171 s |
| Negative framing | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-10-04 22:54 | yes | 15 | 23 s |
Normalization in this category
Every judgment call made between the raw labels and the numbers above, listed so it is visible and reversible.
ShowHide
Category-scoped readings
ABBYY read as ABBYY Vantage
ABBYY (FlexiCapture / Vantage) read as ABBYY Vantage
AWS Textract read as Amazon Textract
Google Cloud Document AI read as Google Document AI
Hyperscience read as Hyperscience Hypercell
Klippa read as Klippa DocHorizon
Klippa (Doxis AI.dp) read as Klippa DocHorizon
Klippa (now Doxis) read as Klippa DocHorizon
UiPath read as UiPath Document Understanding
UiPath IXP read as UiPath Document Understanding
Unresolved, counted raw
ABBYY FineReader Server
Adobe Acrobat
Astera
Azure Content Understanding
BILL AP/AR
DocumentIQ
ExtractBee
Hyland IDP
Indico
Inhubber
Iron Mountain InSight® DXP
KlearStack
Llama Index
Microsoft AI Builder-based workflows
Microsoft document processing / Syntex
OCR.space
OpenText Intelligent Capture
PDF.ai
Paperless-ngx
Parashift
Parsio
Super.AI
Tesseract
Thomson Reuters Document Intelligence
UiPath Agentic Automation
UiPath Agentic Automation / UiPath Platform
Unstructured
Zapier Central
agorum core
Discontinued, still offered
No shut-down product was recommended here.