RxNoteRxNote
    தொடர்பு
    உள்நுழை

    RxNote Read-Bench

    On 130 real handwritten prescriptions from an Indian hospital, Gemini 4 Argon found the most medicines, 82%, ahead of Gemini 3.8 Flash at 79%, which costs about two thirds as much. When two models from different vendors must agree, the list is 93% correct.

    By Amine El Farssi, co-founder and AI lead, RxNote

    Version v2 · run 6 நவம்பர், 2026 · 529 pages · 10 models · 3 tracks

    See the leaderboardRead the method
    Real handwriting: top three130 pages, ranked by medicines found
    1. 1Gemini 4 Argon$8.76 per 1,000 pages82%
    2. 2Gemini 3.8 Flash$5.51 per 1,000 pages79%
    3. 3Claude Opus 5.5$34.33 per 1,000 pages75%

    Bars show 95% intervals over pages. Prices are list prices on the run date; the table shows each price after any introductory offer ends.

    • The prompt
    • Results (JSON)
    • Cite
    • Request the synthetic set

    Leaderboard

    130 pages of handwritten prescriptions and notes from an Indian hospital partner, 739 medicines in the key. Private: we publish scores only.

    Medicines found is the share of the 739 medicines in the reviewed key that the model listed, after dosage form and strength are set aside; a name counts at 85% character similarity or better. Correct when listed is the share of the distinct names the model listed that the key confirms; the rest are misreadings and inventions together. Balanced (F1) combines the two, so a model cannot rank high by listing everything or by listing almost nothing. Bars show 95% intervals over pages.

    Rank by
    1. 1Gemini 4 ArgonGoogleMedicines found81.7%77.9%–85.4%
      Correct when listed
      87.3%83.7%–90.4%87 of 684 unconfirmed
      Balanced (F1)
      84.4%
      $ per 1,000 pages
      $8.76$17.52 after intro
    2. 2Gemini 3.8 FlashGoogleMedicines found79.2%75.3%–82.8%
      Correct when listed
      85.6%82.1%–88.9%97 of 674 unconfirmed
      Balanced (F1)
      82.3%
      $ per 1,000 pages
      $5.51$11.03 after intro
    3. 3Claude Opus 5.5AnthropicMedicines found75.1%70.7%–79.1%
      Correct when listed
      81.5%78%–85%124 of 672 unconfirmed
      Balanced (F1)
      78.2%
      $ per 1,000 pages
      $34.33
    4. 4GPT-6.1 TerraOpenAI (via Amazon Bedrock)Medicines found74.3%69.9%–78.6%
      Correct when listed
      84.3%81%–87.5%101 of 644 unconfirmed
      Balanced (F1)
      79%
      $ per 1,000 pages
      $5.76
    5. 5GPT-6.1 SolOpenAI (via Amazon Bedrock)Medicines found71.9%67.5%–76.2%
      Correct when listed
      79.8%75.5%–83.4%133 of 658 unconfirmed
      Balanced (F1)
      75.6%
      $ per 1,000 pages
      $10.29
    6. 6Gemini 3.1 ProGoogleMedicines found70.4%65.8%–74.9%
      Correct when listed
      77.1%73.1%–80.8%153 of 667 unconfirmed
      Balanced (F1)
      73.6%
      $ per 1,000 pages
      $9.55
    7. 7Claude Sonnet 5.5AnthropicMedicines found68.2%64.3%–72.2%
      Correct when listed
      76.1%72.2%–80.2%156 of 654 unconfirmed
      Balanced (F1)
      72%
      $ per 1,000 pages
      $15.92
    8. 8Gemini 3.5 Flash-LiteGoogleMedicines found64%59.6%–68.4%
      Correct when listed
      70.2%65.5%–74.6%199 of 667 unconfirmed
      Balanced (F1)
      66.9%
      $ per 1,000 pages
      $1.78
    9. 9Claude Haiku 5.5AnthropicMedicines found59.7%55%–64.4%
      Correct when listed
      65.2%60.2%–70.1%233 of 669 unconfirmed
      Balanced (F1)
      62.3%
      $ per 1,000 pages
      $1.03
    10. 10Textract + Claude Haiku 5.5OCR + model · AWS + AnthropicMedicines found50.7%46.6%–55%
      Correct when listed
      60.8%56.4%–65.3%239 of 610 unconfirmed
      Balanced (F1)
      55.3%
      $ per 1,000 pages
      $1.79
    Real handwriting, 130 pages. Select a column heading to sort.
    #Model
    1Gemini 4 ArgonGoogle81.7%95% interval 77.9% to 85.4%87.3%95% interval 83.7% to 90.4%87 of 684 unconfirmed84.4%$8.76$17.52 after intro
    2Gemini 3.8 FlashGoogle79.2%95% interval 75.3% to 82.8%85.6%95% interval 82.1% to 88.9%97 of 674 unconfirmed82.3%$5.51$11.03 after intro
    3Claude Opus 5.5Anthropic75.1%95% interval 70.7% to 79.1%81.5%95% interval 78% to 85%124 of 672 unconfirmed78.2%$34.33
    4GPT-6.1 TerraOpenAI (via Amazon Bedrock)74.3%95% interval 69.9% to 78.6%84.3%95% interval 81% to 87.5%101 of 644 unconfirmed79%$5.76
    5GPT-6.1 SolOpenAI (via Amazon Bedrock)71.9%95% interval 67.5% to 76.2%79.8%95% interval 75.5% to 83.4%133 of 658 unconfirmed75.6%$10.29
    6Gemini 3.1 ProGoogle70.4%95% interval 65.8% to 74.9%77.1%95% interval 73.1% to 80.8%153 of 667 unconfirmed73.6%$9.55
    7Claude Sonnet 5.5Anthropic68.2%95% interval 64.3% to 72.2%76.1%95% interval 72.2% to 80.2%156 of 654 unconfirmed72%$15.92
    8Gemini 3.5 Flash-LiteGoogle64%95% interval 59.6% to 68.4%70.2%95% interval 65.5% to 74.6%199 of 667 unconfirmed66.9%$1.78
    9Claude Haiku 5.5Anthropic59.7%95% interval 55% to 64.4%65.2%95% interval 60.2% to 70.1%233 of 669 unconfirmed62.3%$1.03
    10Textract + Claude Haiku 5.5OCR + model · AWS + Anthropic50.7%95% interval 46.6% to 55%60.8%95% interval 56.4% to 65.3%239 of 610 unconfirmed55.3%$1.79

    Cost against handwriting accuracy

    Each point is one model on the selected track: list price per 1,000 pages on a log scale against handwriting recall, with its 95% interval. The line joins the models no other model beats on both price and recall.

    Real handwriting

    Cost against handwriting recall, Real handwritingScatter plot of 10 models. Frontier models, cheapest first: Claude Haiku 5.5, Gemini 3.5 Flash-Lite, Gemini 3.8 Flash, Gemini 4 Argon.40%45%50%55%60%65%70%75%80%85%90%$1$2$5$10$20$50$ per 1,000 pages (log scale)Medicines foundGemini 3.8 FlashGemini 4 ArgonClaude Opus 5.5GPT-6.1 TerraGPT-6.1 SolGemini 3.1 ProClaude Sonnet 5.5Gemini 3.5 Flash-LiteClaude Haiku 5.5Textract + Claude Haiku 5.5
    Pareto frontier (ringed points)GoogleAnthropicOpenAIAWS Textract + AnthropicIntroductory price, dotted to the price after the introductionThe leaderboard table above holds the same numbers.

    Two readers

    When two models from different vendors must both list a medicine, the list is right 93% of the time, against 86% for the best single model. It catches 68% of the medicines instead of 79%: what only one model saw is what a second look is for.

    Real handwriting, the same 130 pages and key as the leaderboard. Bars show 95% intervals over pages. We do not show 'either model lists it': the key was built from the models' own lists, so that setup would look better than it is.
    SetupCorrect when listedMedicines found
    Gemini 3.8 Flash alone86%95% interval 82% to 89%79%95% interval 75% to 83%
    Gemini 3.8 Flash and GPT-6.1 Terra both list it93%95% interval 90% to 95%68%95% interval 63% to 74%
    Gemini 3.8 Flash and Claude Opus 5.5 both list it93%95% interval 90% to 95%71%95% interval 66% to 75%
    At least two of Gemini 3.8 Flash, Claude Opus 5.5 மற்றும் GPT-6.1 Terra list it91%95% interval 89% to 94%77%95% interval 73% to 81%

    Why the models differ

    On real handwriting, Claude Opus 5.5 misspells medicines about as often as the Gemini models do. It falls behind in two other ways: it leaves more medicines out entirely, and it says when it is unsure.

    Where the misses come from

    Claude Opus 5.5 leaves out 80 medicines with nothing close in its list, against 35 for Gemini 4 Argon, and lists 69 names unlike anything in the key, against 43. Where a scrawl is ambiguous, Gemini more often lands on the right drug; Opus more often names another one or leaves it out.

    ModelMissed, misspeltMissed, nothing close listedListed, close to a key nameListed, unlike any key name
    Gemini 4 Argon100354443
    Gemini 3.8 Flash114405740
    Claude Opus 5.5104805569
    GPT-6.1 Terra92984358

    Real handwriting, 739 medicines in the key. A miss is 'misspelt' when the model listed a name 60–85% similar; below 60% nothing close was listed. Counts, not names: the pages are private.

    Doubt, said out loud

    Our prompt says: never guess at content you cannot read; name the gap instead. Claude Opus 5.5 follows it. It marks 123 of 130 pages as partly unreadable and names 483 unreadable regions; Gemini 3.8 Flash marks 12 and names 23. Its doubt is well placed: names it flags as unsure are right 27% of the time, names it does not flag 86%, close to Gemini 4 Argon at 87%.

    ModelPages marked partly unreadableUnreadable regions namedNames flagged unsureRight when flaggedRight when not flagged
    Gemini 4 Argon42730–87%
    Gemini 3.8 Flash1223250%86%
    Claude Opus 5.51234835627%86%
    GPT-6.1 Terra121285462%90%

    The score counts a flagged name like any other and rewards a right guess, so the leaderboard measures reading, not caution. For a clinician, a model that says where it is unsure is worth something this table does not show.

    Not image resolution

    Every model received the same page images, about 1,744 × 1,241 pixels. Claude Opus 5.5 spent about 3,318 input tokens per page, the Gemini models 1,437 அல்லது 1,536, prompt included. Opus was given at least as much of the image, so a coarser view does not explain its misses.

    Why Textract + Haiku trails Haiku alone

    Textract + Claude Haiku 5.5 runs Amazon Textract on the page and gives Claude Haiku 5.5 only the text it returns; Haiku never sees the image. Handwriting is where OCR breaks down: 28% of Textract's words on these pages came back under 80% confidence, and its text holds only 52% of the medicines in the key. Haiku lists 51% from that text, repairing some garbled names, but it cannot bring back what the text dropped. Reading the page itself, Haiku finds 60%, and 155 of the medicines it finds were nowhere in Textract's text. The pipeline also costs more: $1.79 per 1,000 pages against $1.03.

    • In Textract's text52%
    • Textract + Claude Haiku 5.5 lists51%
    • Claude Haiku 5.5 reading the page60%
    Share of the 739 key medicines. The first bar counts names Textract's text spells within 85% of the key; the pipeline works only from that text, so a few garbled names it repairs are all it can add.

    We tested one OCR engine, Textract's text detection. A stronger handwriting OCR would raise the ceiling, but the limit stays: the second model can only list what the OCR text holds.

    Five words, read three ways

    • Handwritten word labelled Ceftriaxone in the RxHandBD dataset

      Dataset label: Ceftriaxone

      Gemini 4 Argon
      Ceftriaxone
      Gemini 3.8 Flash
      Ceftriaxone
      Claude Opus 5.5
      Cefuroximeflagged unsureflagged unsure

      Opus read a different antibiotic, and flagged it as unsure.

    • Handwritten word labelled Zenup in the RxHandBD dataset

      Dataset label: Zenup

      Gemini 4 Argon
      Zenup
      Gemini 3.8 Flash
      Zenup
      Claude Opus 5.5
      not found

      Opus left this one out.

    • Handwritten word labelled Xinc in the RxHandBD dataset

      Dataset label: Xinc

      Gemini 4 Argon
      Xinc
      Gemini 3.8 Flash
      xinc
      Claude Opus 5.5
      Zincflagged unsureflagged unsure

      Opus read the everyday word, Zinc, and flagged it.

    • Handwritten word labelled Deixil in the RxHandBD dataset

      Dataset label: Deixil

      Gemini 4 Argon
      Deanxit
      Gemini 3.8 Flash
      Denixit
      Claude Opus 5.5
      Derixil

      Both Gemini models read a different, real drug name. The handwriting itself reads closer to Derixil than to the label.

    • Handwritten word labelled Nexe in the RxHandBD dataset

      Dataset label: Nexe

      Gemini 4 Argon
      Nene
      Gemini 3.8 Flash
      Nene
      Claude Opus 5.5
      Nexeflagged unsureflagged unsure

      Opus read it, and flagged it as unsure; both Gemini models read Nene.

    From the public synthetic set: handwriting from RxHandBD (© Md. Masudul Islam, MIT licence), cut from the pages exactly as the models saw them. The real pages stay private. Across the set's 1,560 handwritten words, both Gemini models read 76 that Opus did not; Opus read 30 that neither Gemini model did.

    What we think explains the rest

    • Brand familiarity. Indian prescriptions are written in brand names. With three or four legible letters, a model that knows the brand list can complete the word; the Gemini models' wrong readings tend to be close spellings, Opus's more often a different word.
    • Instruction following. A prompt without the 'never guess' line might let Opus guess more and find more. We have not run it; the prompt stays the same for every model.

    Both are hypotheses, not results.

    Five things a clinician needs from a reader

    • Handwriting recall

      Every handwritten word on the page, medicines first, against a word-level answer key. A word counts when the reading matches it at 85% character similarity or better, so a one-letter slip in a long name passes and a different drug does not.

    • Printed fields

      Medicines, doses, lab values and dates, including pages printed in Hindi and Tamil. A lab value counts only when the number is exactly right.

    • Honesty

      On a torn or smudged page the model should name the gap, not fill it. A blank page should come back empty. A chart, stamp or photo should be flagged as an image, not transcribed as text.

    • Reliability

      Of the drug names a model lists, how many are really on the page, and lab numbers that differ from the printed value. A model that lists everything it might see finds more and is right less often, so we rank on both and on their balance.

    • Cost and latency

      List price per 1,000 pages on the run date, from each vendor’s published rates and the tokens the call used, and the median seconds per page.

    How the sets are built

    • Synthetic set

      Generated prescriptions and lab reports in Indian layouts, with real handwritten medicine names from RxHandBD placed into them. RxHandBD is openly available on Zenodo. The generator, the pages and the answer key are shared with research teams on request.

      Handwriting from Islam, Md. Masudul (2026). RxHandBD. Zenodo. doi:10.5281/zenodo.18478741

    • Private real sets

      Lab reports and handwritten prescriptions from an Indian hospital partner, used with consent. The pages are never released and never leave our evaluation environment; we publish scores only.

    • Answer key for real handwriting

      Pages were chosen by a neutral rule: Amazon Textract's handwriting tag, at least 12 handwritten words, spread across source documents. Every model's medicine names were pooled and shown blind to a human reviewer, without model names and in shuffled order. The reviewer confirmed, corrected or rejected each one and added the medicines no model found. 20 of 150 pages had no readable medicine and were set aside. Gemini 4 Argon was run after this review, so the 32 names only it listed were reviewed the same way afterwards; the 4 the reviewer could not decide count against it. The key was built by one reviewer, blind to model names; a second, independent reviewer is planned for v3.

    Methodology and limitations

    Methodology

    • Every model gets the same prompt and must answer in the same JSON contract.
    • Vendor-default settings: no tuning per model. A page whose call failed (timeout, throttling, a reply that broke the JSON contract) was run once more; a page that failed again stays a failure and is shown in the table.
    • One call per page. A reply that breaks the contract scores zero on that page and is counted as a failure.
    • 95% intervals come from a bootstrap over pages. Where we claim one model beats another, we use a paired bootstrap over the same pages: Gemini 3.8 Flash leads Claude Opus 5.5 by 4.0 points of medicine recall on the real pages (95% interval 1.0 to 7.0) and by 3.5 points of handwriting recall on the synthetic set (2.0 to 5.1).
    • Cost is the tokens each call used at the vendor's list price on the run date, averaged over the synthetic set. Gemini 3.8 Flash is shown at its introductory price, with the list price after the introduction. Amazon Bedrock reports almost no input tokens for the GPT-6.1 models, so we price their input with OpenAI's published image-token formula for a page this size.

    Limitations

    • Synthetic pages are cleaner than real scans: no folds, shadows or carbon copies. The real tracks score different things (medicines on a page, lab values), so their numbers are not comparable with the synthetic handwriting recall.
    • RxHandBD words are Bangladeshi brand names. They test handwriting, but they are not the Indian brands a prescription here would carry.
    • The private sets are small: 130 handwritten and 71 lab pages from one hospital. Read the intervals before reading the ranks.
    • The real-handwriting key starts from what the models read, and the reviewer adds what they all missed. A medicine every model and the reviewer missed is not in the key.
    • Conflict of interest: RxNote builds products on some of these models. We are not paid by any vendor.
    The prompt every model receivedSent with one page image per call, unchanged for every model and every page.
    You are transcribing one page of a patient's medical record for a clinician.
    Return ONLY a JSON object with these keys:
    - "markdown": the full page content as markdown in reading order. Tables as markdown tables. Preserve every drug name, dose, frequency, lab value, unit, date and code verbatim. Transcribe handwriting as best you can. Describe any chart, photo, stamp or signature in one line in square brackets, e.g. [chart: fasting glucose trend, values 120, 135, ...].
    - "language": "en", "hi", "ta", "mixed" or "none".
    - "legibility": "readable" if you could read everything, "partial" if some region could not be read, "unreadable" if most of the page could not be read, "empty" if the page is blank.
    - "unreadable_regions": list of short descriptions of any region you could NOT read reliably (blurred, blacked out, illegible handwriting). Empty list if none. Never guess at content you cannot read; name the gap instead.
    - "medications": list of {"name","dose","frequency"} found on the page (strings, verbatim).
    - "labs": list of {"analyte","value","unit"} found on the page.
    - "dates": list of dates found, as written.
    - "handwritten_words": list of the handwritten words you transcribed, in order (empty if no handwriting).
    - "image_described": true if the page contains a chart, photo, stamp or signature that you described.
    No prose outside the JSON.
    Cite this benchmarkBibTeX for the version on this page.
    @misc{rxnote_readbench_v2_2026,
      title        = {RxNote Read-Bench v2: how well AI models read Indian medical records},
      author       = {El Farssi, Amine and {RxNote}},
      year         = {2026},
      howpublished = {\url{https://rxnote.ai/benchmark}},
      note         = {Version v2, run 2026-11-06}
    }

    Run it on your own model

    We can score your model on the private sets without the pages leaving our environment, and share the synthetic set, its generator and its answer key with research teams.

    Score your model on the private sets

    RxNote LogoRxNote

    உங்கள் ஏஐ பராமரிப்பு கூட்டாளர்

    பின்வரும் புதுமை திட்டங்களால் ஆதரிக்கப்படுகிறது

    Microsoft Logo
    Nvidia Logo
    HIPAA தயார்நிலைGDPR தயார்நிலை

    தயாரிப்பு

    • Intake
    • Scribe
    • Ward Notes
    • Assistant
    • Document Analyser
    • Dictation
    • Template Builder
    • Care Journey

    Specialties

    • All Specialties
    • Psychiatry
    • Family Medicine
    • Internal Medicine
    • Cardiology
    • Pediatrics
    • Mental Health
    • Emergency Medicine

    Use Cases

    • All Use Cases
    • SOAP Notes Generator
    • Pre-Visit Prep
    • Follow-Up Letters
    • Telemedicine
    • Solo Practice
    • Clinic/Hospital

    வளங்கள்

    • எங்களை பற்றி
    • வலைப்பதிவு
    • Customer Stories
    • Benchmark
    • Data & Security
    • ICD-11 Lookup
    • FAQs
    • வேலை வாய்ப்புகள்

    © 2025 RxNote Medical AI. அனைத்து உரிமைகளும் பாதுகாக்கப்பட்டவை.

    தனியுரிமை கொள்கைபயன்பாட்டு விதிமுறைகள்முகவர் Markdown
    RxNote