Benchmark v1.0 · report

dsh/rtx5090/qwen3.8-27b-nvfp4

setup

model type: open model (local)
hardware: RTX5090
harness: dsh
model: qwen3.8-27b-nvfp4

modelself-reportedqwen38-27b-nvfp4-128k

started 2026-10-06 10:11 UTC

overall

Answered 27 of 49 challenges; 26 correct.

26 of 49 challenges passed

partial run · 22 unanswered, counted against the score

26 passed
1 failed
22 not answered

vitals

time

1h 18m

answered

55%

failed

success

53%

systems

Math test

9/9 passed

1m 30s

letter-count-1✓ pass52s
prompt
How many times does the letter "x" appear in "zanmonxixx"? Answer with just the number.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
Trivial. I counted x manually and double-checked with a script; both gave 3. No ambiguity in the prompt.
decimal-compare-1✓ pass4s
prompt
Which decimal number is larger, 3.89 or 3.2? Answer with just the larger number.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
Straightforward decimal comparison; 3.89 > 3.2. Routine, no hesitation.
arithmetic-1✓ pass5s
prompt
Compute step by step, left to right (no operator precedence): 14 + 20 / 2 * 3 * 7. Answer with just the final number.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
The explicit left-to-right instruction is a trap for habit: with normal precedence it would be a different value. I followed the stated order exactly (34, 17, 51, 357) and verified with a script. Confident in 357.
unit-convert-1✓ pass5s
prompt
Convert 16 km to m. Now treat that resulting number as a fresh quantity of km and convert it to m (1 km = 1000 m). Answer with just the final integer number of m.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
Two chained conversions: 16 km -> 16000 m, then 16000 treated as km -> 16000000 m. Easy once you parse the odd second step of treating meters as km.
format-json-1✓ pass5s
prompt
Reply with ONLY a JSON object, no other text. The object must have exactly two keys, in this order: "answer" then "checksum". "answer" must be the string "5085". "checksum" must be a JSON number (not a string) equal to the sum of the digits of 5085. Example shape: {"answer":"1234","checksum":10}
answer
answer hidden on shared reports
grader · score 1
agent's debrief
Formatting test more than math: keys in exact order, checksum as a JSON number not a string, no surrounding text. 5+0+8+5=18. I kept the payload pure JSON as required.
math-add-1✓ pass5s
prompt
What is 11 + 13? Answer with just the number.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
Trivial addition, 11+13=24. Instant.
math-sum-1✓ pass5s
prompt
Compute the exact sum of 328 + 796. Respond with just the integer.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
Simple two-digit-plus two-digit addition; 328+796=1124, verified with a script.
math-expression-1✓ pass4s
prompt
Evaluate this expression exactly and respond with just the integer: ((-19 + -17) * (35 - 27)) + (-1 * -1) - 33
answer
answer hidden on shared reports
grader · score 1
agent's debrief
The negative signs are the hazard: (-36)*8 + 1 - 33 = -320. Computed by hand and confirmed with a script; confident.
math-determinant-1✓ pass5s
prompt
Compute the exact determinant of this 4x4 matrix. Respond with just the integer. [9, 6, -2, -1] [-2, 2, 0, -8] [-7, -8, -5, 9] [-7, 5, -7, -3]
answer
answer hidden on shared reports
grader · score 1
agent's debrief
The hardest of the section. I expanded along the row with a zero by hand, then cross-checked with exact fraction-based elimination in a script; both gave -4214. Comfortable in the answer.

Vision test

17/19 passed · 1 unanswered

1h 18m

acuity-20✓ pass2m 07s
prompt
Look at the image at (fetch it and view it). This is an eye chart with 7 numbered rows, each containing 3 groups of 5 characters. Read row 4, group 2 (groups are numbered left to right). What are the 5 characters, in order? Answer with just the 5-character code, e.g. AB3XZ.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
I fetched the eye chart and read row 4, group 2: PCNFT | WQCEF | RP72K. The characters were legible at this size; confident in WQCEF.
acuity-14✓ pass8s
prompt
Look at the image at (fetch it and view it). This is an eye chart with 7 numbered rows, each containing 3 groups of 5 characters. Read row 5, group 1 (groups are numbered left to right). What are the 5 characters, in order? Answer with just the 5-character code, e.g. AB3XZ.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
Same kind of eye-chart read, now at a smaller row: row 5 group 1 shows AC7Q8. I zoomed mentally on the smaller text; the Q is a bit soft but the grouping makes it unambiguous. Reasonably confident.
acuity-10✓ pass35s
prompt
Look at the image at (fetch it and view it). This is an eye chart with 7 numbered rows, each containing 3 groups of 5 characters. Read row 6, group 3 (groups are numbered left to right). What are the 5 characters, in order? Answer with just the 5-character code, e.g. AB3XZ.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
Row 6 is small on the full chart, so I installed Pillow, cropped just that row and upscaled 2x before reading it. VNH7R is clear in the crop. The first character is V not U, based on the pointed baseline.
acuity-8✓ pass25s
prompt
Look at the image at (fetch it and view it). This is an eye chart with 7 numbered rows, each containing 3 groups of 5 characters. Read row 7, group 1 (groups are numbered left to right). What are the 5 characters, in order? Answer with just the 5-character code, e.g. AB3XZ.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
The smallest row on the chart; I had to crop just group 1 of row 7 and upscale 8x to read it reliably. 4YF8V looks unambiguous at that zoom.
count-simple✓ pass28s
prompt
Look at the image at (fetch it and view it). How many purple triangles are in the image? Answer with just the number.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
Three purple triangles visible, and I cross-checked with a pixel-level connected-component count (3 purple blobs, triangle-sized). Confident.
count-medium✓ pass30s
prompt
Look at the image at (fetch it and view it). How many purple circles are in the image? Answer with just the number.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
I classified all 17 purple connected components by fill-ratio: 14 round ones (fill ~0.77) plus one diamond and two triangles (fill ~0.51), which are not circles. My visual count (3+2+5+4) matched. Confident in 14.
count-complex✓ pass43s
prompt
Look at the image at (fetch it and view it). How many green triangles are in the image? Answer with just the number.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
Too many shapes to count reliably by eye. I ran connected-component analysis on the green colour: 38 green blobs, then classified each by fill-ratio and top/bottom row width — 24 triangles, 7 circles, 2 squares, 3 diamonds. The shape separation was clean (triangles: top point width 2, full-width base), so I am confident in 24.
spatial-simple✓ pass15s
prompt
Look at the image at (fetch it and view it). Which cell contains the red circle? Answer as: row R, column C (1-indexed from top-left).
answer
answer hidden on shared reports
grader · score 1
agent's debrief
Straightforward grid read: the only red circle sits in the bottom row, fourth cell. Unambiguous. Used the answer format given in the prompt.
spatial-medium✓ pass6m 16s
prompt
Look at the image at (fetch it and view it). Which shape has an arrow pointing to the orange square? Answer with just the colour and shape, e.g. teal diamond.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
The arrowhead at the orange square comes from the line running up-right to the blue triangle in row 5 col 3. A second line runs FROM the orange square down to the blue diamond, and the green circle points at the green square, so no confusion. Confident.
spatial-complex✕ fail38m 10s
prompt
Look at the image at (fetch it and view it). Which shape is 3 steps before the orange circle along the arrows? Answer with just the colour and shape, e.g. teal diamond.
answer
answer hidden on shared reports
grader · score 0
wrong shape
agent's debrief
Traced all 15 arrows by programmatic dark-pixel component analysis with 9x9 density-based arrowhead detection, then zoomed crops to confirm the green-circle/green-triangle/white-square region. The unique chain into the orange circle is green triangle -> green circle -> red diamond -> green circle... i.e. three hops back lands on the red diamond.
chart-simple✓ pass3m 42s
prompt
Look at the image at (fetch it and view it). Using the "New Signups" chart, approximately what value did Feb have? Read it off the y-axis; answers within +/-5 are accepted.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
Bar chart of New Signups; the Feb bar sits just above the 30 gridline, read as 33 within the +/-5 tolerance.
chart-medium✓ pass1m 27s
prompt
Look at the image at (fetch it and view it). Using the "Monthly Active Users" chart, approximately what value did Aug have? Read it off the y-axis; answers within +/-5 are accepted.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
MAU bar chart in thousands; the Aug bar reaches the midpoint between the 70 and 80 gridlines, read as 75 thousand within +/-5.
chart-complex✓ pass2m 14s
prompt
Look at the image at (fetch it and view it). Using the "Support Tickets Opened" chart, how many months did New have a value greater than 71? Answer with just the number.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
Grouped bar chart, New (blue) vs Returning (orange) per month. New exceeds 71 only in Jun (~88), Oct (~84), Nov (~82), Dec (~95); Aug (~65) and Sep (~61) stay below. Count = 4.
screenshot-simple✓ pass3m 34s
prompt
Look at the image at (fetch it and view it). What is the total amount shown in the cart panel? Answer with just the total amount, e.g. $12.34.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
Cart screenshot: Water Bottle x3 @ $45.99 = $137.97 plus Mouse Pad x2 @ $8.07 = $16.14; line totals and grand total $154.11 all check out arithmetically.
screenshot-medium✓ pass2m 30s
prompt
Look at the image at (fetch it and view it). What is the total amount shown in the cart panel? Answer with just the total amount, e.g. $12.34.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
Five-line cart; recomputed each line total (32.37, 16.26, 11.43, 32.02, 44.12) and summed to $136.20, matching the displayed Total.
screenshot-complex✓ pass2m 44s
prompt
Look at the image at (fetch it and view it). What is the line total for Keyboard on this order summary? Answer with just the amount, e.g. $12.34.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
Order summary with ten line items; the Keyboard row is x2 at $19.96 unit, line total $39.92, which I verified by multiplication.
diagram-simple✓ pass2m 59s
prompt
Look at the image at (fetch it and view it). This diagram shows named boxes connected by arrows. Which box has an arrow pointing to "Ridge"? Answer with just the box name, e.g. Kettle.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
Flow diagram: Piano branches to Kiwi and Valley; Valley is the only box with an arrow into Ridge.
diagram-medium✓ pass9m 39s
prompt
Look at the image at (fetch it and view it). This diagram shows named boxes connected by arrows. Which box has an arrow pointing to "Zebra"? Answer with just the box name, e.g. Kettle.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
Two crossing edges on the right of the middle layer: Tunnel line goes up-right into Zebra, Opals lower line crosses it to Jetty; verified via 2.5x zoomed crop.
diagram-complex— unanswered—
prompt
Look at the image at (fetch it and view it). This diagram shows named boxes connected by arrows. Which box has an arrow pointing to "Rocket"? Answer with just the box name, e.g. Kettle.

Finding and reading email test

not examined · 0/6 answered

aggregate-1— unanswered—
prompt
You are examining a mailbox: Phillip Allen's mail at enronmail.airbench.ai How many messages carry the label "markets"? Answer with just the number.
aggregate-2— unanswered—
prompt
You are examining a mailbox: Phillip Allen's mail at enronmail.airbench.ai How many messages include jwills3@swbell.net in the To field? Answer with just the number.
temporal-1— unanswered—
prompt
You are examining a mailbox: Phillip Allen's mail at enronmail.airbench.ai What is the subject of the newest message in the sent folder? Answer with just the subject line, exactly as shown.
temporal-2— unanswered—
prompt
You are examining a mailbox: Phillip Allen's mail at enronmail.airbench.ai What is the subject of the oldest message in the archive folder? Answer with just the subject line, exactly as shown.
needle-1— unanswered—
prompt
You are examining a mailbox: Phillip Allen's mail at enronmail.airbench.ai In the message with the subject "FW: Mid C New deals Sept 24", what net value to book is given for the new deals? Answer with just the number.
needle-2— unanswered—
prompt
You are examining a mailbox: Phillip Allen's mail at enronmail.airbench.ai In Julieta Sandoval's message about the Muni Bond Ladder, what direct phone number does she give? Answer with just the exact fact as it appears in the message, and nothing else.

Purchasing test

not examined · 0/4 answered

find-product-1— unanswered—
prompt
The store is at abostore.airbench.ai Among products in the **Pet Supplies** category priced under **$400** with a rating of at least **4**, which has the **lowest price**? The search space is every product in the store's catalog. Answer with just that product's id (the amazon.xx:B0… form).
find-product-2— unanswered—
prompt
The store is at abostore.airbench.ai Among products in the **Sports & Outdoors** category priced under **$800** with a rating of at least **4.5**, which has the **lowest price**? The search space is every product in the store's catalog. Answer with just that product's id (the amazon.xx:B0… form).
purchase-1— unanswered—
prompt
The store is at abostore.airbench.ai Buy exactly 3 units of AmazonBasics Enamel Cast Iron Kettlebell - 25 Pounds, Black (product id amazon.ca:B0731CP5S4, abostore.airbench.ai/product/amazonbasics-enamel-cast…) on abostore, completing checkout with a valid payment method. Use the checkout email aidoctor-f4ed05d0@aidoctor.test. Answer with just the resulting order id.
recover-decline-1— unanswered—
prompt
The store is at abostore.airbench.ai Buy exactly 1 unit of Simple Joys by Carter's Baby Girl's Hat and Mitten Set (product id amazon.co.uk:B07BBJGRCM, abostore.airbench.ai/product/simple-joys-by-carter-s-…) on abostore. First attempt checkout with a card number ending in 0000 — this payment will be declined. Then retry checkout with a different, valid card and complete the purchase successfully. Use the same checkout email aidoctor-2c372133@aidoctor.test for both attempts. Answer with just the order id of the successful (approved) order.

Coding test

not examined · 0/11 answered

compute-hash-1— unanswered—
prompt
Write and run a short program to compute this exactly. Use unsigned 32-bit arithmetic modulo 2^32 throughout. Define rotl32(z, r) = ((z << r) | (z >> (32 - r))) mod 2^32, and let imul(a, b) mean 32-bit multiplication modulo 2^32. Start with data = [73321616, 1150564817, 2160567878, 949665919, 173444172, 560083901, 3263843362, 1338515211, 1925302088, 606482921, 2483508030, 1655513047], x = 908138884, y = 3339223125 and rounds = 25000. For each step from 0 to rounds - 1 apply: x = (imul(x XOR data[step % 12] XOR step, 2654435761) + rotl32(y, 5)) mod 2^32; y = (imul(y + data[(step + 7) % 12] + step, 1597334677) XOR rotl32(x, 11)) mod 2^32; x = (x + rotl32(y XOR step, 3)) mod 2^32. Return the final value as two lowercase 8-digit hex words joined by a hyphen, like 1a2b3c4d-55667788.
compute-vm-1— unanswered—
prompt
Write and run a program to execute this tiny machine exactly. It has four registers a, b, c, d, all starting at 0, and a program counter starting at line 0. `set r n` sets r to n. `add r x`, `sub r x` and `mul r x` combine r with x (a register name or an integer), then reduce r modulo 1000003 into the range 0..1000002. `dec r` subtracts 1 from r. `jnz r k` jumps k lines (relative; k may be negative) if r is not 0. `halt` stops. Every other instruction moves to the next line. 0: set a 55 1: set b 951 2: set c 255 3: set d 572 4: add b a 5: mul a 92 6: mul b 18 7: dec d 8: jnz d -4 9: add a 42 10: dec c 11: jnz c -8 12: halt Respond with just the final value of register a.
compute-paths-1— unanswered—
prompt
Here is a grid of 25 rows and 25 columns. S is the top-left start, E is the bottom-right end, # is a wall and . is open. You may move up, down, left or right onto any non-wall cell. Find the length of the shortest path from S to E (the number of moves) and the number of distinct shortest paths modulo 1000000007. Write a program; this is not meant to be done by eye. S.#.............#.#.#.#.# .......##..#..........##. .##..#....##...#...#.#.#. #.....#..#......#.##...#. .#...#....##.#......#.##. ...#..#.#.#....##..#..#.. #.#.#.#.#.#........#..... .....#..#.#.##.....##.... .#.#........#...##....... #...#.##......#...#...### #..###.#......###.#....#. .#.#.#....#....#...#.###. #...#.......###.#..#...#. #.#....#...#......##...## .#.......#.......##...#.. ..#....#...#.#........... .#...#.#..#..#.......#..# ##...###.#.....###.##.#.. #.#.............#..#..#.# ...#....#...##...#....... ##.##..##.#.#..##.###.##. ...#.....#.#...##...##.## ..##.............#....... #..##.#.#...#.##......... ##..#....#..#...##.....#E Respond with the two integers separated by a space, like `52 1840`.
compute-life-1— unanswered—
prompt
Simulate Conway's Game of Life on a 20x20 grid that wraps around at every edge (a torus). # is alive and . is dead. A live cell with 2 or 3 live neighbours (out of its 8) stays alive; a dead cell with exactly 3 live neighbours becomes alive; every other cell is dead in the next generation. Starting grid, row 0 first: .###.##.#...#.#..... .#.#.........#..#... ...#....#...###.##.# .........##.....#.#. #..##.......#.##.#.# .#.......##.....###. ##......#.#......... ##...#####.#......#. #.#.#...###..##.##.. #..#...........#.... #..###...###....##.. #.#...#.....##.##..# #..#..##.###.#....#. .###.#..........##.. .#......#..#..#.##.. #.....#...#....#.... .#..#..#..#.####..#. ##..###.##........#. #..#.#..##......#.## .....#...#.#..#..... Run 150 generations. Report the number of live cells and the sum of row*20+column over all live cells (rows and columns numbered from 0). Respond as live:sum, like `37:7421`.
compute-fibmod-1— unanswered—
prompt
Let F(0) = 0, F(1) = 1 and F(k) = F(k-1) + F(k-2). Compute F(n) mod m exactly for n = 6262788590067066 and m = 999983. Respond with just the integer.
compute-words-1— unanswered—
prompt
Below is a text. Words are separated by spaces. Ignore letter case, and strip any punctuation or quote characters attached to a word. Count how often each word occurs, then report the 3 most frequent words, most frequent first, breaking ties alphabetically. Respond exactly as word=count,word=count,word=count. renvo Quiti kasha Mobas Zanzan QUILU rensha ficfic. DORLU; ficvo renvo kasha mozan renvo; kavo? Dorlu dordor QUITI kavo kasha! ficbas, kavo dorsha kabas dorsha dorsha karen Quilu quipel kavo! karen ficbas, ficfic. Quipel. quipel kavo zantru zanzan, ficbas moka Quipel quiti nixren trumo mobas mopel, dordor ficfic? dordor zantru quiti rensha shabas, SHABAS. Quilu kabas ficbas dordor dorsha kasha kasha dordor Nixsha Kabas kasha renvo "Zantru" ficbas nixren QUIPEL Quipel zanpel quilu Kasha Ficfic Nixren quipel ficbas quipel Quilu KASHA Kavo "shabas" ficvo kasha dorlu "Trumo" "quipel" mozan shatru ficbas! renvo Zanpel shazan Karen quiti Ficbas Shazan Quipel Quiti ficbas ficbas! "Dorsha" dorsha! mozan Shabas shazan Shatru quipel motru quipel quiti moka kasha dorlu Zantru Ficbas rensha ficbas kasha shatru mobas quiti dorlu ficfic zantru ficbas zanzan kasha Ficvo moka dorsha shatru renvo, "kasha" "Kasha" dorsha kavo shatru moka? quipel renvo? dorlu nixsha rensha nixsha, "quipel" kasha renvo dordor? Dorsha KASHA ficbas Zanzan dordor Mobas dorlu Quipel nixren quipel Nixsha renvo rensha quipel quipel; Mobas ficvo renvo motru. Quipel Renvo. ficbas dorsha mozan; dordor ZANZAN Zanpel Kasha zanzan shazan timo ficbas renvo zantru dorlu dorsha zantru dorlu quipel "ficvo" Quipel zantru! renvo, nixren Zantru nixren Zanpel renvo mobas Nixsha motru Quipel! Timo Renvo Quiti Shazan Renvo mobas mozan zantru Shatru ficvo ficfic kasha? Quiti? zanpel quipel mozan; renvo MOBAS quipel zantru renvo motru dorlu. dorlu KAVO ficvo Zanpel dorlu Quipel zantru moka Mobas! renvo shazan rensha Moka ficvo; "dorlu" Nixsha dorsha. ficvo dorsha quiti nixsha kabas Zantru Renvo "dorsha" kabas nixsha Ficvo Nixsha karen shatru kasha. quipel. Shatru quilu Shazan dorlu zanpel Zantru kavo; dordor zanpel zanpel dordor moka motru ficfic renvo quipel mozan moka kabas zanzan dorsha Karen Kavo kasha mozan. quipel Zantru dorsha kasha zanzan renvo kasha zanpel karen. renvo shabas dorlu kavo mozan! ficbas, nixren timo zantru kasha kasha zantru renvo Shazan ficbas quilu zanpel zanpel "dorsha" Renvo shatru mozan Ficbas renvo ficbas quipel! "nixren" quilu dorlu shabas kasha quipel! Mopel nixsha nixren; zanpel ficbas karen mopel MOKA Mobas kasha zantru quipel Renvo Kasha mozan "Kavo" ficbas ficbas shazan Moka Mozan quiti nixsha dorsha ficbas quipel! QUITI mopel moka kasha zantru kasha motru renvo quipel mozan? "kavo" mopel Ficvo quipel ficbas kabas dorsha Dordor; dorsha trumo ficbas. trumo Nixsha Shazan Quiti Kavo kasha zanpel. dordor quiti renvo quiti karen "ficbas" kasha nixsha? ficbas dordor Mozan nixren Renvo; trumo Shabas TRUMO renvo, kasha "Renvo" Karen quipel; ficvo Zanpel Kasha kasha shabas! quiti dordor dorsha zanzan renvo, nixren, Dorsha nixsha quipel mopel KABAS dorsha mozan dorlu quipel Zanzan
trace-1— unanswered—
prompt
What exactly does this JavaScript program print? Respond with just the printed output. const v1 = ["30" < "4", null >= 0, null == 0].map(Number).join(""); const v2 = ["6", "85", "101"].map(parseInt).join(","); const v3 = (0.1 * 1 + 0.2 * 1 === 0.3 * 1) ? "equal" : "different"; const v4 = [18 / 6 | 0, Math.round(-6.5), -97 % 2].join(","); console.log(v1, v2, v3, v4);
fix-1— unanswered—
prompt
Our shipping quote function below has one bug. Bug report: this order is quoted 975 cents, but the correct quote is 490: {"country":"FR","items":[{"grams":1726,"qty":1,"price":4500,"fragile":false}]} Fix the bug without changing any other behaviour, then run the fixed quote() on each of the 20 orders below, in order. Respond with just the 20 results separated by commas. const ZONES = { FR: 1, DE: 1, ES: 1, IT: 1, GB: 2, US: 2, CA: 2, JP: 3, BR: 3, AU: 3 }; // any other country is zone 4 const BASE = [0, 485, 860, 1325, 1641]; // cents, by zone const PER_STEP = [0, 70, 148, 225, 260]; // cents per 250 g step, by zone const FREE_BASE_OVER = [0, 4500, 10300, 15000, Infinity]; // order value (cents) that waives the base fee function quote(order) { const zone = ZONES[order.country] ?? 4; let grams = 0; let value = 0; let fragile = 0; for (const item of order.items) { grams += item.grams * item.qty; value += item.price * item.qty; if (item.fragile) fragile += item.qty; } const steps = Math.max(1, Math.ceil(grams / 250)); let cents = PER_STEP[zone] * steps; if (value <= FREE_BASE_OVER[zone] || order.express) cents += BASE[zone]; cents += Math.min(fragile, 3) * (120 + 35 * zone); if (order.express) cents = Math.ceil((cents * (zone <= 2 ? 150 : 185)) / 100); if (order.coupon === "SHIP10") cents -= Math.min(cents >> 3, 500); return Math.max(cents, 99); } Orders: {"country":"US","items":[{"grams":974,"qty":1,"price":4234,"fragile":false},{"grams":1316,"qty":5,"price":7003,"fragile":false}]} {"country":"CA","items":[{"grams":263,"qty":2,"price":391,"fragile":true},{"grams":1575,"qty":1,"price":1119,"fragile":true},{"grams":933,"qty":2,"price":2622,"fragile":false}],"express":true} {"country":"DE","items":[{"grams":1555,"qty":1,"price":2475,"fragile":false},{"grams":467,"qty":1,"price":7419,"fragile":false}]} {"country":"ES","items":[{"grams":1583,"qty":1,"price":6815,"fragile":false}]} {"country":"MX","items":[{"grams":327,"qty":2,"price":4927,"fragile":false},{"grams":899,"qty":4,"price":6782,"fragile":true},{"grams":1130,"qty":5,"price":8330,"fragile":true}]} {"country":"BR","items":[{"grams":1288,"qty":1,"price":2423,"fragile":false},{"grams":1281,"qty":1,"price":7621,"fragile":false},{"grams":1325,"qty":3,"price":6740,"fragile":false}],"coupon":"SHIP10"} {"country":"DE","items":[{"grams":380,"qty":1,"price":4500,"fragile":false}]} {"country":"GB","items":[{"grams":1560,"qty":1,"price":8647,"fragile":true},{"grams":1022,"qty":1,"price":8084,"fragile":false},{"grams":852,"qty":5,"price":662,"fragile":true},{"grams":898,"qty":1,"price":8251,"fragile":false}]} {"country":"JP","items":[{"grams":433,"qty":2,"price":4035,"fragile":false},{"grams":674,"qty":1,"price":419,"fragile":true}],"express":true} {"country":"NZ","items":[{"grams":1676,"qty":1,"price":4371,"fragile":false},{"grams":1300,"qty":2,"price":7058,"fragile":false},{"grams":99,"qty":4,"price":1063,"fragile":false},{"grams":917,"qty":2,"price":5485,"fragile":false}],"coupon":"SHIP10"} {"country":"CA","items":[{"grams":755,"qty":2,"price":6434,"fragile":false},{"grams":667,"qty":1,"price":7164,"fragile":false}]} {"country":"IT","items":[{"grams":1934,"qty":1,"price":4500,"fragile":false}]} {"country":"JP","items":[{"grams":1493,"qty":1,"price":15000,"fragile":false}]} {"country":"BR","items":[{"grams":1542,"qty":4,"price":6891,"fragile":false},{"grams":1532,"qty":1,"price":2277,"fragile":true},{"grams":421,"qty":1,"price":1637,"fragile":false},{"grams":143,"qty":1,"price":6949,"fragile":false}],"express":true} {"country":"GB","items":[{"grams":1873,"qty":1,"price":10300,"fragile":false}]} {"country":"BR","items":[{"grams":392,"qty":1,"price":15000,"fragile":false}]} {"country":"CA","items":[{"grams":598,"qty":1,"price":1588,"fragile":false}],"express":true} {"country":"BR","items":[{"grams":143,"qty":3,"price":648,"fragile":false},{"grams":311,"qty":3,"price":2888,"fragile":true}],"express":true} {"country":"GB","items":[{"grams":1140,"qty":1,"price":10300,"fragile":false}]} {"country":"JP","items":[{"grams":1056,"qty":1,"price":15000,"fragile":false}]}
implement-1— unanswered—
prompt
Write a function mergeIntervals(intervals). Each interval is a pair [start, end] of integers with start <= end, and includes both ends. It returns the intervals with every overlapping or touching pair merged (so [1, 3] and [3, 5] become [1, 5], but [1, 2] and [3, 4] stay apart), sorted by start. The input may be in any order. Run it on each of the 12 inputs below and respond with one line per input: the result as JSON, like [[1,5],[8,10]]. Spaces are ignored. [] [[1,3],[3,5]] [[1,2],[3,4]] [[5,9],[1,10]] [[9,15],[38,41],[9,17]] [[39,44],[3,3],[9,15],[18,23],[27,33],[29,37],[9,9],[24,32]] [[21,25],[34,37],[11,15],[7,7]] [[31,35],[4,9],[36,41],[13,19]] [[16,17],[24,31],[40,45],[11,16],[40,41],[24,24]] [[36,36],[28,34],[11,16],[18,26],[9,15]] [[35,35],[35,36],[3,4],[19,20],[17,20],[30,32],[24,25],[3,7]] [[20,25],[8,12],[15,18],[17,19],[36,41]]
repo-1— unanswered—
prompt
Download airbench.ai/f/8856175e276baa07fe02db1849150502.zip and unzip it. It contains a small Python project (Python 3.10 or newer, standard library only) with a bug. Find and fix it, then run `python main.py data/real.csv` inside the project folder and respond with just the 8-character code it prints.
repo-2— unanswered—
prompt
Download airbench.ai/f/cc994cc2674aa994b7f89694413fc08b.zip and unzip it. It contains a small Python project (Python 3.10 or newer, standard library only) with bugs. Find and fix them, then run `python main.py data/real.csv` inside the project folder and respond with just the 8-character code it prints.

how this agent was configured

Hardware: NVIDIA RTX 5090 32 GB (SM120), x86_64, driver 595.84, Ubuntu 24.04 (host hal5090). Model server: Checkpoint gittensor-model-hub/Qwen3.8-27B-NVFP4-RTX5090 (NVFP4, no MTP head). vLLM 0.27.1 (vllm/vllm-openai:v0.27.1): --quantization modelopt --kv-cache-dtype fp8 --trust-remote-code --reasoning-parser qwen3 --enable-auto-tool-choice --tool-call-parser qwen3_xml --max-model-len 131072 --max-num-seqs 4 --gpu-memory-utilization 0.95. ~79 tok/s single-stream decode. Harness: dsh 0.2.0-rc.2 (DeepSeek Harness, @deepseek-ai/dsh) in a container (node:22): `dsh --profile headless --patch <file> --json <prompt>` with DSH_PERMISSION_MODE=danger-full-access and DSH_TELEMETRY_MODE=DISABLED; the patch adds one pi-ai openai-completions route; context 131072, max output 32768 tokens and makes it the default model, same overlay as gx10's dsh agent; everything else, compaction included, is the headless profile's default. Orchestrator: github.com/dh7/agent-checkup-benchmark @ 19423df; the checkup instructions are passed verbatim as a single prompt, one attempt, no time limit unless noted. Model requests pass through the orchestrator's local proxy, which only intervenes when the server rejects prompt + max_tokens as over the context window: it then retries once with max_tokens lowered to fit. Operator limits: stopped by the operator at 120 min (120-min cap).