AI Demos Research — the structured-intelligence platform. Every verdict on these pages opens to the execution behind it.
Graded 30 October 2026

Does GPT-5.6 terra keep equations as math when converting a PDF to Markdown?

GPT-5.6 terra kept the document's equations as math markup rather than flattening them into prose. The source note had six numbered equations plus one unnumbered relation, and the returned Markdown carried all seven as LaTeX with numbering preserved; the title, byline, sections, body prose, and references also carried over.

1 of 1 test case passed

Every test case in this scenario has a result.

Pass rate100%1 of 1 gradable
Coverage1 of 1test case with a result
1PassDid everything it was expected to do.
0FailNo test case failed.
0Not gradableEvery result here could be graded.
0UntestedEvery test case in this scenario has a result.
The pass rate is a summary. The evidence is the test case below: what we sent, what we checked, what the tool returned, and the proof.

The test case

Each test case is judged on its own: Pass, Fail, Not gradable, or Untested. The scenario result above is based on this test case.

Equations — math markup or honest fallbackPassEvidence
What we sent
InputDocument (PDF)
Input PDF with six equations and one unnumbered relation
Open the PDF ↗
What the tool returned
Markdownthe whole file · cited lines 1–93 and 101–106 marked
# A one-dimensional model of tidal attenuation in the Braithe channel

Technical note TIH/TN/18 · January 2026

R. K. Sandiman and H. Vale

## 1 Introduction

The tide entering the Braithe is attenuated over the eleven kilometres between the bar and Braithe Bridge, and the attenuation is large enough to matter to anyone predicting water levels in the upper estuary. This note sets out the one-dimensional model the Institute has used since 2018, states the approximation on which the working formula rests, and compares the result with the gauge record at three stations.

Nothing here is new. The intention is to have the derivation, the assumptions and the coefficients in one place, because the formula has been quoted in Partnership papers without them.

## 2 Governing equations

Take the channel as a single reach of slowly varying cross-section and neglect lateral inflow. Conservation of mass gives

$$
\frac{\partial A}{\partial t}
+
\frac{\partial Q}{\partial x}
= 0
\tag{1}
$$

where $A$ is the cross-sectional area below the free surface and $Q$ the discharge through it. Conservation of momentum, with friction represented in the Manning form, gives

$$
\frac{\partial Q}{\partial t}
+
\frac{\partial}{\partial x}
\left(
\frac{Q^2}{A}
\right)
+ gA
\frac{\partial \eta}{\partial x}
+
\frac{g n^2 Q |Q|}{A R^{4/3}}
= 0
\tag{2}
$$

in which $\eta$ is the surface elevation above mean sea level, $n$ is Manning's coefficient, $R$ the hydraulic radius for wetted perimeter $P$, and $g$ the acceleration due to gravity. The friction term is the only nonlinear one that matters over the range of interest.

$$
R = A/P
$$

## 3 Attenuation of the leading harmonic

For a channel of nearly uniform depth the leading semidiurnal harmonic decays close to exponentially with distance upstream, so that

$$
\eta(x, t) = \eta_0 e^{-\mu x}\cos(\omega t - kx)
\tag{3}
$$

with $\eta_0$ the amplitude at the bar, $\omega$ the angular frequency of the harmonic, $k$ the wavenumber and $\mu$ the attenuation coefficient. Linearising the friction term by the Lorentz method and collecting terms gives

$$
\mu =
\frac{\omega}{c}
\left(
\frac{\chi}{2(1 + \chi^2)}
\right)^{1/2},
\qquad
\chi =
\frac{8 n^2 c \eta_0}{3\pi h^{4/3}}
\tag{4}
$$

where $c$ is the frictionless wave speed and $h$ the mean depth. The second expression is the friction parameter; it is small in the lower reach and approaches unity above Harrowby Reach, which is where the exponential form begins to fail.

## 4 Application to the Braithe

Taking the mean depth as eleven metres below the bar and four metres at the bridge, with Manning's coefficient at the value fitted in 2018, the formula reproduces the observed amplitude at Vardenne Quay to within four per cent and at Harrowby Reach to within nine. At Braithe Bridge it overestimates the amplitude, and the discrepancy grows through the spring tides, which is the behaviour the friction parameter predicts.

A two-reach treatment with separate depths would probably remove most of the error at the bridge. It has not been attempted here because the gauge record above Harrowby Reach is too short to fit a second coefficient with any confidence.

## 5 Coefficients

Manning's coefficient is not constant with depth in a channel of this kind, and the single fitted value used above is a compromise. Fitting the 2018 record with a depth-dependent form gives

$$
n(h) = n_\infty +
\frac{b}{h^{1/3}}
\tag{5}
$$

with the deep-water value $n_\infty$ and $b$ a bed constant. The fit is better in the lower reach and no better at the bridge, which suggests that the error there is not in the friction term at all.

The two constants were fitted to the 2018 record and have not been refitted since. Refitting on the four years now available would be worth doing; the Institute's expectation is that the deep-water value will move very little and the bed constant appreciably, because it is carrying the shallow reaches where the record has improved most.

## 6 Limitations

Three limitations should be stated plainly. The first is the one-dimensional assumption, which fails where the channel divides above Harrowby Reach. The second is the neglect of the freshwater inflow, which is small in summer and not small after rain. The third is the linearisation itself, which is a poor approximation once the friction parameter approaches unity - that is, in exactly the reach where the model is least accurate.

The error measure quoted in the previous section is

$$
\varepsilon =
\frac{|\eta_o - \eta_m|}{\eta_o}
\tag{6}
$$

taken over the spring-neap cycle at each gauge, with the observed amplitude $\eta_o$ and the modelled one $\eta_m$. It is a crude measure and it flatters the model at the bar, where the amplitude is large and the absolute error is not.

None of this is an argument against using the formula, which is quick, transparent and good enough for the purposes the Partnership puts it to. It is an argument for quoting it with the reach and the tidal range attached, which the papers that quote it have not always done.

## References

Sandiman, R. K. (2018). *Tidal propagation in the Braithe: a first fit.* Tarnbeck Institute internal report TIH/IR/09.

Vale, H. and Croyde, P. (2021). *The Harrowby Reach gauge, 2015 to 2020.* *Journal of Estuarine Hydrology* 44, 218-31.

Prosser, A. (2009). *Friction coefficients for the Braithe channel.* Braithe Harbour Board, unpublished.

Markdown output with all equations as LaTeX math

Copied from Output file (Markdown), lines 1–93, 101–106 · an excerpt

Open the whole Markdown file ↗
Expected vs. Found
✓Expected: Equations should stay as math markup or an honest fallback, not turn into garbled prose.Found: All seven equations appear as LaTeX, with numbering preserved where present, and none are garbled or missing.
Supporting proof
Proof 1Screenshot
Page 1 comparison of source and Markdown
Open original ↗
Proof 2Screenshot
Page 2 comparison of source and Markdown
Open original ↗
Proof 3Screenshot
Page 3 comparison of source and Markdown
Open original ↗
Proof 4Screenshot
Run card with completion, timing, and settings
Open original ↗
Proof 5Captured file (JSON)
API response with returned Markdown and settings
Cited values7 values from this file
model
gpt-5.6-terra
temperature
1
reasoning
{ "context": "all_turns", "effort": "medium", "mode": "standard", "summary": null }
service_tier
default
output 2 › content 1 › text
# A one-dimensional model of tidal attenuation in the Braithe channel Technical note TIH/TN/18 · January 2026 R. K. Sandiman and H. Vale ## 1 Introduction The tide entering the Braithe is attenuated over the eleven kilometres between the bar and Braithe Bridge, and the attenuation is large enough to matter to anyone predicting water levels in the upper estuary. This note sets out the one-dimensional model the Institute has used since 2018, states the approximation on which the working formula rests, and compares the result with the gauge record at three stations. Nothing here is new. The intention is to have the derivation, the assumptions and the coefficients in one place, because the formula has been quoted in Partnership papers without them. ## 2 Governing equations Take the channel as a single reach of slowly varying cross-section and neglect lateral inflow. Conservation of mass gives $$ \frac{\partial A}{\partial t} + \frac{\partial Q}{\partial x} = 0 \tag{1} $$ where $A$ is the cross-sectional area below the free surface and $Q$ the discharge through it. Conservation of momentum, with friction represented in the Manning form, gives $$ \frac{\partial Q}{\partial t} + \frac{\partial}{\partial x} \left( \frac{Q^2}{A} \right) + gA \frac{\partial \eta}{\partial x} + \frac{g n^2 Q |Q|}{A R^{4/3}} = 0 \tag{2} $$ in which $\eta$ is the surface elevation above mean sea level, $n$ is Manning's coefficient, $R$ the hydraulic radius for wetted perimeter $P$, and $g$ the acceleration due to gravity. The friction term is the only nonlinear one that matters over the range of interest. $$ R = A/P $$ ## 3 Attenuation of the leading harmonic For a channel of nearly uniform depth the leading semidiurnal harmonic decays close to exponentially with distance upstream, so that $$ \eta(x, t) = \eta_0 e^{-\mu x}\cos(\omega t - kx) \tag{3} $$ with $\eta_0$ the amplitude at the bar, $\omega$ the angular frequency of the harmonic, $k$ the wavenumber and $\mu$ the attenuation coefficient. Linearising the friction term by the Lorentz method and collecting terms gives $$ \mu = \frac{\omega}{c} \left( \frac{\chi}{2(1 + \chi^2)} \right)^{1/2}, \qquad \chi = \frac{8 n^2 c \eta_0}{3\pi h^{4/3}} \tag{4} $$ where $c$ is the frictionless wave speed and $h$ the mean depth. The second expression is the friction parameter; it is small in the lower reach and approaches unity above Harrowby Reach, which is where the exponential form begins to fail. ## 4 Application to the Braithe Taking the mean depth as eleven metres below the bar and four metres at the bridge, with Manning's coefficient at the value fitted in 2018, the formula reproduces the observed amplitude at Vardenne Quay to within four per cent and at Harrowby Reach to within nine. At Braithe Bridge it overestimates the amplitude, and the discrepancy grows through the spring tides, which is the behaviour the friction parameter predicts. A two-reach treatment with separate depths would probably remove most of the error at the bridge. It has not been attempted here because the gauge record above Harrowby Reach is too short to fit a second coefficient with any confidence. ## 5 Coefficients Manning's coefficient is not constant with depth in a channel of this kind, and the single fitted value used above is a compromise. Fitting the 2018 record with a depth-dependent form gives $$ n(h) = n_\infty + \frac{b}{h^{1/3}} \tag{5} $$ with the deep-water value $n_\infty$ and $b$ a bed constant. The fit is better in the lower reach and no better at the bridge, which suggests that the error there is not in the friction term at all. The two constants were fitted to the 2018 record and have not been refitted since. Refitting on the four years now available would be worth doing; the Institute's expectation is that the deep-water value will move very little and the bed constant appreciably, because it is carrying the shallow reaches where the record has improved most. ## 6 Limitations Three limitations should be stated plainly. The first is the one-dimensional assumption, which fails where the channel divides above Harrowby Reach. The second is the neglect of the freshwater inflow, which is small in summer and not small after rain. The third is the linearisation itself, which is a poor approximation once the friction parameter approaches unity - that is, in exactly the reach where the model is least accurate. The error measure quoted in the previous section is $$ \varepsilon = \frac{|\eta_o - \eta_m|}{\eta_o} \tag{6} $$ taken over the spring-neap cycle at each gauge, with the observed amplitude $\eta_o$ and the modelled one $\eta_m$. It is a crude measure and it flatters the model at the bar, where the amplitude is large and the absolute error is not. None of this is an argument against using the formula, which is quick, transparent and good enough for the purposes the Partnership puts it to. It is an argument for quoting it with the reach and the tidal range attached, which the papers that quote it have not always done. ## References Sandiman, R. K. (2018). *Tidal propagation in the Braithe: a first fit.* Tarnbeck Institute internal report TIH/IR/09. Vale, H. and Croyde, P. (2021). *The Harrowby Reach gauge, 2015 to 2020.* *Journal of Estuarine Hydrology* 44, 218-31. Prosser, A. (2009). *Friction coefficients for the Braithe channel.* Braithe Harbour Board, unpublished.
usage
{ "input_tokens": 1605, "input_tokens_details": { "cache_write_tokens": 1602, "cached_tokens": 0 }, "output_tokens": 1732, "output_tokens_details": { "reasoning_tokens": 382 }, "total_tokens": 3337 }
status
completed
Show the whole file (11 KB)
JSON
{
  "id": "resp_0b23ac4c7000a9b4006ab372eb6a6887d1bc7996764f5333e3",
  "object": "response",
  "created_at": 1790145259,
  "status": "completed",
  "access_programs": null,
  "background": false,
  "billing": {
    "payer": "developer"
  },
  "completed_at": 1790145273,
  "error": null,
  "frequency_penalty": 0.0,
  "incomplete_details": null,
  "instructions": null,
  "max_output_tokens": null,
  "max_tool_calls": null,
  "model": "gpt-5.6-terra",
  "moderation": null,
  "output": [
    {
      "id": "rs_0b23ac4c7000a9b4006ab372ec40dc87d1a8555bbb89fe3591",
      "type": "reasoning",
      "content": [],
      "encrypted_content": "gAAAAABqs3L56N5YVjP1mxCLs9Hht2BpCAbQzsRr9QAEZqGEeHWCYqrDZYlEB8HCfCQh-7arocgcQ38okK_RyOp6m2NhhOZmt1mtgtP0ietYFUssMbEGWrMNawAWfGaTTVFPzz4EkRwoM1qmc2a0qkFRsVlPfZg8kpd6IdwfN6KobzKIFUUBUN2F_8H3oVyKU_nLX1yaHFnWrT0cIqtbiI3byOLAUuU0lZ6RHEWlFBufcwwZ2aCjVCQLBc9wf37A4R6_UmteZRdRKEyFsBOLK4vi8AlXM2GYqcpJBrXoKn3mZNbYZkCqtXmEs4eImFthvFv24WhjW_MqZ8h0eP1JVHSf3heGXZgocHMdCKB6Yna5pP07gi63o5HbvUZivMANWsM5BM1A163CJvBuuNIjRGrMri3rcFpL0N-mbX6et9B_7yE_jFxn6DPFAJ_W0Xf6ep95ZwHBhT0uLnIaZfeBXSdn6l0i5dHJKRvFNa5I5853e11YQAgIVYoc1VpAAsVYEtriX8LD6FVu4kWsY5-OO0ga0hzY7V1sjRPqfQZr-TrJ8Ayzikpkl1NJ2X-tLQefTNvjvAjkyjaRWl22PabYFVA0jL9LLZOdrrprf3p7H0r3iNnTKqPR6WPBwddkOFpWrEDWDvUz8afCRw5sBcZtM5ASoMen6vECcL3MUXz7Kq6UdafmdxrOt8WvIwzVgk9tyx3dZZXd6rLzY2TJFCga_PVnWeEYDWZjTEE0cu6JD2XvZ62WjeAPT8xdGqUu88yusJ4ZJj-qWsGSOd-_GhunwV2uFpntNGyCbpi-dP21SD8sMDQHaAtyF64cuZUwcqfyiX13axRzQIjQOGJcPVGIfDUt14u_pH8YsG70eIbiPZ34uZJgc56DHjzJvczVVUbGrWvYKMozMpomIFtcKDyC2rnrlySZ0UFtZ-zvFegF-o9674CoOpcQBOXmTmhuOmDXo065QnQWdCpDbL_Dsn5DaXFV0XyGG0k7xdij3Jl8F5ykrloccljP6JgxDshevGrjTJ7gSRc0_nhK1sGG6VKMCwIi9t6ByRkd1PZJrTO55p-yUa44nAeNAGX47WlsWZ6LWN5EppkTalLxNsyuXI3zpVUQ2cvHR0ePP6p_WcKOJUbY6WDyK8D4ogucruoHvRaERFPjehCNdUs1Jx-upw12YQQhAM45RU_6BH3yJ-w9e3Nl8zaZPeyn5waA_IHDD5nH8uILJbTiTkmAzP3iLAiFinwRPBURoiE--4D3Jv33J3K9e9FT_U4BG0r2En9mtUQEP6NRHM7Oausn7npXU_tGPkL7PZ3_BsJuvIH0ZpnFAGVewzoHMNDAH6iDBiieFQ-5Ux7qAuZJ1DcPYm2IIY9zoZmhaA27XGISEU52ojNo-bPKpnPwaxAj1ilr7oK-dk7Rtq27jyYyb2ndLEuhZB2mdj6bN80piS3uWE70DgLk0E8Bj81BglL5fNoEmi87rbgKzXjEZa9gIVkjm9tvl_WlIo26FwzDKDPx9sgijoEHHsDjvBGUlrjMAh7lWYdIJKB_JtAIYV2XfCdUFvvNENFlAVZsbwTQPAbwa2Er39bgC_khuHxITeW9p8QCLZGWsxU_NyDUf1-thJB-4r0jL5wFJ_DQ12hJl7blBRRBvdstZLV7jHcrEGtZsHnIFa_xWBDCmlWnfcZl5G5wFOu2E2yJP89_eCgvasG6SKa7ReQ6wv1W_FQcDjArsIaQMxzml8EUjJCA98IUGe1Stav6xjOtOTOvxNSVdHEqh97FlNqhGyaLPeZQWK6Mo7xYD3EGap-gw0B8yK6Qd3wQjMKM_UtwN5bOjokkmbkp01_V3xNNxVOohlFJoToZSrqp46cpw7WFezWQMRt3DvCiuIvuzoT4N7Z-XjGMeT7o7Dr9Z7-ClGKImPZNplVf1GqNM4BhyOuBGYp3m-JsDCpZ-oV9XjOWbRZ_zppFoHTA9KQLQBcTf5MzUHokaB0esrjH8Ed_FSHjdcum07dCqT4LE8QTPsHU6NpVmI0nysbhRk8umH9f4GetZA_wSsL-tHKHOMzGbKXKutN4jqkevIxvczbZ_VZbmi0mOGY5-v2DLwMd1Wa2Jb64aKSlVfPtlbjdL-oBsZ3MN_FGQG3sgHKJYO0PxGJjoiyizqtdlprSusMGpsKJ36QU5shX2DXOl4VPoZtjBKUyxNAaNEl-rVQ1EfAXyWXGoqs6QjYwfl2ydXXuV6Naac11yMN4GsHj9rKS9DaxnXeA6ciD4gVFtNaALoG76XOEQBiqxCpyCnZf9GX3toggsypRslYzhTKRvySg8wEX8V3gIBVahdMQA1XLE4kLYKoiTRH0GHF_1P9pnCrxig-ajCGt8CKQh_FPKV85qxd-4C-A12JdLyFfq9437Lc4JtyInwdxcOJbqoTyCkAAtO_hibby8O2ZoTN8NzYcfsXC5x_FuOOlBuKehrwqhasTHHpfjapsBq-LiU-IXaIXas2mDo8a9jq4gdDxTyeOi3axsagIQ8G4lf-QTjA9aH3QXhcnwUAw0-uzG-Iy7dhxV0JFeF3Z0q7SVVmoP6W8YX1wvYjoeau6G-gTyRenRA6fCcPkdDl2tPYKgqL4DvrxcC101w4DimLasGGpn7H_GhT3iJ-Bw4MUNE8SfPWRSiolgvlIFB4NHkYc8EslFkKvu8Np3RIbe99QzZAKWdZ_2iumPh2eH0gt6u0TF4V-bU_HyNP7hSqaP8fOPWd1QXUOKRk4Dye2uDNxsjptmKDYBSRm1h7Oz4vnl-QXqm4c6Av-SXVEZS9FDEpm30wpD3RZs-T5BWa56x3tUH6doBQBBWZZ_up3FMo9wC4CKByPsZnLsPWVMJdpQimwZSqXmh6qRLoySD1276KFC4pQxzcKrDzlA6UjreBs4fjfB2lxix7TZzXFIZ-fipOxj_EnRw8SRbu0YlYZT0E7VVfEneD1VR88BuskcpX-xzUXp-ss7z7-Z88TEdPfdGImAoCKJ_ti7WvK4lt7j3EFd4n5tKoz9MIKwwLQpXFb_eMPAP5cMq3Kbi-GrgjB1pCQxk5Zjnr1XLaMf8CrUXxYnZZtAMRQKLfzHlQsHOlSLMG-eoLszB4dCcNvq9J1EJ5Bnvf_UJwb3btnlCfMquR4y5AyeAjU1E7zwo_Pd7I60cNUI_8ta16SZmUQKEBg8BBvl9X_Gd9V6iRi6ZpBtovBdmhQb2pUqipWVDy4hI1W7A9laOMqbR5PzzE1rKm2C_pWy27pX_kYkdhM32UUduCiqRfnZ5q8mawl3yvGaeyd3FJJljvyMpmL64TGPmfE-Ofo7tno1_GhN7UgjqaG87MFYLnOxtYmvBkIXLlBA0X6GrqUq20zRhT4s-IjkmL1p91HK5fa-Fy8RVTmKmGfM9V4XA6g3OpkKUV3HBWg2J9zDT7UiOTqwsCRfUq69xfNpfwT_a1Q0_6t2UnHd-VkdXFlcLvd-96D6eiJYY0yHt3-8MhcLhlTARdnD_vLSY_GjqrfuKr5CL88BA0HFkKorhuSeokc960=",
      "summary": []
    },
    {
      "id": "msg_0b23ac4c7000a9b4006ab372f17a6487d18e87f84966751488",
      "type": "message",
      "status": "completed",
      "content": [
        {
          "type": "output_text",
          "annotations": [],
          "logprobs": [],
          "text": "# A one-dimensional model of tidal attenuation in the Braithe channel\n\nTechnical note TIH/TN/18 \u00b7 January 2026\n\nR. K. Sandiman and H. Vale\n\n## 1 Introduction\n\nThe tide entering the Braithe is attenuated over the eleven kilometres between the bar and Braithe Bridge, and the attenuation is large enough to matter to anyone predicting water levels in the upper estuary. This note sets out the one-dimensional model the Institute has used since 2018, states the approximation on which the working formula rests, and compares the result with the gauge record at three stations.\n\nNothing here is new. The intention is to have the derivation, the assumptions and the coefficients in one place, because the formula has been quoted in Partnership papers without them.\n\n## 2 Governing equations\n\nTake the channel as a single reach of slowly varying cross-section and neglect lateral inflow. Conservation of mass gives\n\n$$\n\\frac{\\partial A}{\\partial t}\n+\n\\frac{\\partial Q}{\\partial x}\n= 0\n\\tag{1}\n$$\n\nwhere $A$ is the cross-sectional area below the free surface and $Q$ the discharge through it. Conservation of momentum, with friction represented in the Manning form, gives\n\n$$\n\\frac{\\partial Q}{\\partial t}\n+\n\\frac{\\partial}{\\partial x}\n\\left(\n\\frac{Q^2}{A}\n\\right)\n+ gA\n\\frac{\\partial \\eta}{\\partial x}\n+\n\\frac{g n^2 Q |Q|}{A R^{4/3}}\n= 0\n\\tag{2}\n$$\n\nin which $\\eta$ is the surface elevation above mean sea level, $n$ is Manning's coefficient, $R$ the hydraulic radius for wetted perimeter $P$, and $g$ the acceleration due to gravity. The friction term is the only nonlinear one that matters over the range of interest.\n\n$$\nR = A/P\n$$\n\n## 3 Attenuation of the leading harmonic\n\nFor a channel of nearly uniform depth the leading semidiurnal harmonic decays close to exponentially with distance upstream, so that\n\n$$\n\\eta(x, t) = \\eta_0 e^{-\\mu x}\\cos(\\omega t - kx)\n\\tag{3}\n$$\n\nwith $\\eta_0$ the amplitude at the bar, $\\omega$ the angular frequency of the harmonic, $k$ the wavenumber and $\\mu$ the attenuation coefficient. Linearising the friction term by the Lorentz method and collecting terms gives\n\n$$\n\\mu =\n\\frac{\\omega}{c}\n\\left(\n\\frac{\\chi}{2(1 + \\chi^2)}\n\\right)^{1/2},\n\\qquad\n\\chi =\n\\frac{8 n^2 c \\eta_0}{3\\pi h^{4/3}}\n\\tag{4}\n$$\n\nwhere $c$ is the frictionless wave speed and $h$ the mean depth. The second expression is the friction parameter; it is small in the lower reach and approaches unity above Harrowby Reach, which is where the exponential form begins to fail.\n\n## 4 Application to the Braithe\n\nTaking the mean depth as eleven metres below the bar and four metres at the bridge, with Manning's coefficient at the value fitted in 2018, the formula reproduces the observed amplitude at Vardenne Quay to within four per cent and at Harrowby Reach to within nine. At Braithe Bridge it overestimates the amplitude, and the discrepancy grows through the spring tides, which is the behaviour the friction parameter predicts.\n\nA two-reach treatment with separate depths would probably remove most of the error at the bridge. It has not been attempted here because the gauge record above Harrowby Reach is too short to fit a second coefficient with any confidence.\n\n## 5 Coefficients\n\nManning's coefficient is not constant with depth in a channel of this kind, and the single fitted value used above is a compromise. Fitting the 2018 record with a depth-dependent form gives\n\n$$\nn(h) = n_\\infty +\n\\frac{b}{h^{1/3}}\n\\tag{5}\n$$\n\nwith the deep-water value $n_\\infty$ and $b$ a bed constant. The fit is better in the lower reach and no better at the bridge, which suggests that the error there is not in the friction term at all.\n\nThe two constants were fitted to the 2018 record and have not been refitted since. Refitting on the four years now available would be worth doing; the Institute's expectation is that the deep-water value will move very little and the bed constant appreciably, because it is carrying the shallow reaches where the record has improved most.\n\n## 6 Limitations\n\nThree limitations should be stated plainly. The first is the one-dimensional assumption, which fails where the channel divides above Harrowby Reach. The second is the neglect of the freshwater inflow, which is small in summer and not small after rain. The third is the linearisation itself, which is a poor approximation once the friction parameter approaches unity - that is, in exactly the reach where the model is least accurate.\n\nThe error measure quoted in the previous section is\n\n$$\n\\varepsilon =\n\\frac{|\\eta_o - \\eta_m|}{\\eta_o}\n\\tag{6}\n$$\n\ntaken over the spring-neap cycle at each gauge, with the observed amplitude $\\eta_o$ and the modelled one $\\eta_m$. It is a crude measure and it flatters the model at the bar, where the amplitude is large and the absolute error is not.\n\nNone of this is an argument against using the formula, which is quick, transparent and good enough for the purposes the Partnership puts it to. It is an argument for quoting it with the reach and the tidal range attached, which the papers that quote it have not always done.\n\n## References\n\nSandiman, R. K. (2018). *Tidal propagation in the Braithe: a first fit.* Tarnbeck Institute internal report TIH/IR/09.\n\nVale, H. and Croyde, P. (2021). *The Harrowby Reach gauge, 2015 to 2020.* *Journal of Estuarine Hydrology* 44, 218-31.\n\nProsser, A. (2009). *Friction coefficients for the Braithe channel.* Braithe Harbour Board, unpublished."
        }
      ],
      "phase": "final_answer",
      "role": "assistant"
    }
  ],
  "parallel_tool_calls": true,
  "presence_penalty": 0.0,
  "previous_response_id": null,
  "prompt_cache_key": null,
  "prompt_cache_retention": "24h",
  "reasoning": {
    "context": "all_turns",
    "effort": "medium",
    "mode": "standard",
    "summary": null
  },
  "safety_identifier": null,
  "service_tier": "default",
  "store": true,
  "temperature": 1.0,
  "text": {
    "format": {
      "type": "text"
    },
    "verbosity": "medium"
  },
  "tool_choice": "auto",
  "tool_usage": {
    "image_gen": {
      "input_tokens": 0,
      "input_tokens_details": {
        "image_tokens": 0,
        "text_tokens": 0
      },
      "output_tokens": 0,
      "output_tokens_details": {
        "image_tokens": 0,
        "text_tokens": 0
      },
      "total_tokens": 0
    },
    "web_search": {
      "num_requests": 0
    }
  },
  "tools": [],
  "top_logprobs": 0,
  "top_p": 0.98,
  "truncation": "disabled",
  "usage": {
    "input_tokens": 1605,
    "input_tokens_details": {
      "cache_write_tokens": 1602,
      "cached_tokens": 0
    },
    "output_tokens": 1732,
    "output_tokens_details": {
      "reasoning_tokens": 382
    },
    "total_tokens": 3337
  },
  "user": null,
  "metadata": {}
}
Open the JSON file ↗
Why this result

The document had to keep equations as math. The returned Markdown shows every equation as LaTeX and the full document text also carries over, under the complete PDF sent in one default pass, so this is a Pass.

Also observed in this test case

Faithful beyond equations. The title, byline, section headings, body prose, and references all carried over, and the output stopped where the source stopped.

Tested by Chandresh Bisht · evidence dated 23 September 2026

Configuration and setup

How this tool was set up for the run and what the test needed in place. Each row is a fact from the run's records; a fact the records do not hold is left out, not guessed.

Software that produced the output
GPT-5.6 terra
Build or version
gpt-5.6-terra (model id echoed by the API)
Plan or tier
OpenAI API, credit-funded key (reference subject)
Surface
API
Set up before the run
A PDF with mathematical equations had to be converted to Markdown in one default pass, using the complete file and no per-tool tuning.
Reasoning effort
medium
Text verbosity
medium
Truncation
off
Conversion instruction version
pdf2md-v1
Run setup
no tools, single pass
Tested
23 September 2026 · Chandresh Bisht

How this scenario is graded

How we decide Pass, Fail and Not gradable. The same rules apply to every tool tested on this scenario.

How results are decided

Each test case gets one result per tool: Pass, Fail or Not gradable. A test case we haven't run yet shows Untested. There are no partial results.

The rules
  • Pass: the tool did everything the test expected, and nothing it said contradicts the correct answer.
  • Fail: at least one expected behaviour clearly didn't happen; the test case's result says which one and quotes the tool.
  • Not gradable: our evidence couldn't settle the outcome (for example a record we needed is missing). It is never counted as a fail, and the test case's result says what's missing.

Where this sits in the benchmark

This page is one cell of a larger study: one tool, one scenario. Only this benchmark's frame appears here.

History of this result

What has happened to this result since it was first published. Runs and grades are never overwritten: a retest or a re-grade publishes a new result and keeps the earlier one readable.

from the publication record
18 November 2026First publishedConverting a complex PDF into clean Markdown with a hosted API v1

Act on this result

Nothing filed here edits the run or the grade. A challenge opens a review, and a review can produce a new run or a re-grade — which becomes the current result and leaves this one in the history.

This matches what I see

You run the same kind of test against your own setup and get the same behaviour.

Agree →
This does not match

Yours behaves differently. Tell us what you got, with a screenshot if you have one.

Disagree →
Point out an issue

Something here is wrong — a reference value, a transcription, a grade.

Report an issue →
Request a retest

On a newer build, a larger dataset, or your own setup.

Request a retest →
We have fixed this

Tell us what changed and we schedule a rerun of the failing test case. The old result stays as history.

Vendor notice →
Filed against this evidenceNothing yet. Challenges, counter-evidence and fix notices appear here with their outcome, and stay on the page after they are resolved.
Cite this result
aidemos.com/benchmarks/pdf-to-markdown-apis/results/gpt-5-6-terra/a-document-containing-mathematical-equations · 1 pass · 0 fail · coverage 1/1 · graded 2026-10-30

The same record is available as structured data through the AI Demos MCP server, with the counts, the coverage and every per-test-case reason carried as fields.

Verify the proof files

These files support this result. Open a file to inspect the original evidence.

Input · Document (PDF) ↗application/pdf · 236 KB
Output file (Markdown) ↗text/markdown · 5 KB
Proof 1 · Screenshot ↗image/png · 464 KB
Proof 2 · Screenshot ↗image/png · 494 KB
Proof 3 · Screenshot ↗image/png · 207 KB
Proof 4 · Screenshot ↗image/png · 93 KB
Proof 5 · Captured file (JSON) ↗application/json · 11 KB
File fingerprints (SHA-256)

A fingerprint identifies the exact file used for this result.

Input · Document (PDF)f3d69b6d9d2191c5806a426fd4ea5e209dcd7cf693ad6abe67a5d18a014b9fd6
Output file (Markdown)d7fc5d8e264e37f6af5a1ff00188d28aa260fbc6b849be49c5089140213d1ba8
Proof 1 · Screenshot8ba73e1c40f139a51f3bcf0ec4b2fd67b930abfcfe7da054622f8a1dfe7b39f8
Proof 2 · Screenshot63a7cc93662ec1a8de32a485fd2b60b298e7ee35c53df541e38c18707ea745df
Proof 3 · Screenshot3d3adb0b1294003d0cef8d16ae9f0a979f92ca293205dd9a344c73726890091e
Proof 4 · Screenshotb94c09192b70eef77f546ec0f3369d49a0561bafe00e71ba8277b0ed2afa2e39
Proof 5 · Captured file (JSON)9e7eacbbbe21aeb9d8857112470ab867faa5617eae8a0e476be9adee773e190e