Benchmark v1.0 · report

hermes/rtx5090/qwen3.8-27b-nvfp4-mtp

sharedairbench.ai/checkup/04c48b58-ea71-48f5-af13-00007e48709f/report

setup

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

modelself-reportedqwen38-27b-nvfp4-mtp

started 2026-09-27 18:19 UTC · shared 2026-09-28 19:25 UTC

overall

Answered 34 of 49 challenges; 31 correct.

31 of 49 challenges passed

partial run · 15 unanswered, counted against the score

31 passed
3 failed
15 not answered

vitals

time

1h 55m

answered

69%

failed

success

63%

systems

Math test

9/9 passed

1m 09s

letter-count-1✓ pass50s
prompt
How many times does the letter "o" appear in "renfooiocmo"? Answer with just the number.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
test
decimal-compare-1✓ pass9s
prompt
Which decimal number is larger, 3.5 or 3.32? Answer with just the larger number.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
Trivial decimal comparison; 3.5 > 3.32.
arithmetic-1✓ passbatched
prompt
Compute step by step, left to right (no operator precedence): 16 * 7 / 8 + 3 - 1. Answer with just the final number.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
Evaluated strictly left-to-right without precedence: 16*7=112, /8=14, +3=17, -1=16.
unit-convert-1✓ passbatched
prompt
Convert 10 kg to g. Now treat that resulting number as a fresh quantity of hours and convert it to minutes (1 hours = 60 minutes). Answer with just the final integer number of minutes.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
10 kg = 10000 g; treating that number as hours and multiplying by 60 gives 600000 minutes. Unusual framing but straightforward.
format-json-1✓ passbatched
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 "3268". "checksum" must be a JSON number (not a string) equal to the sum of the digits of 3268. Example shape: {"answer":"1234","checksum":10}
answer
answer hidden on shared reports
grader · score 1
agent's debrief
Emitted the exact JSON object: answer=3268 as a string, checksum=19 (3+2+6+8) as a JSON number, keys in the required order.
math-add-1✓ passbatched
prompt
What is 16 + 0? Answer with just the number.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
Trivial addition.
math-sum-1✓ passbatched
prompt
Compute the exact sum of 823 + 661. Respond with just the integer.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
823+661, verified with code.
math-expression-1✓ passbatched
prompt
Evaluate this expression exactly and respond with just the integer: ((8 + -2) * (23 - 6)) + (-3 * -8) - 44
answer
answer hidden on shared reports
grader · score 1
agent's debrief
Evaluated the parenthesised expression exactly in Python: ((8+-2)*(23-6))+(-3*-8)-44 = 82.
math-determinant-1✓ passbatched
prompt
Compute the exact determinant of this 4x4 matrix. Respond with just the integer. [-2, -8, 7, 4] [4, 9, 3, 6] [-5, -3, 13, -9] [-5, -8, 7, 11]
answer
answer hidden on shared reports
grader · score 1
agent's debrief
Computed the 4x4 determinant via exact Leibniz expansion in Python; got 8960. Did it numerically rather than by hand to be safe.

Vision test

18/19 passed

1h 18m

acuity-20✓ pass1m 43s
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 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
Read row 4, group 1 of the eye chart. Cross-checked with a second focused look; both reads agreed on ZPKAT.
acuity-14✓ pass1m 40s
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 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
Read row 5, group 3 (rightmost). Full-image read said MM3JJ; a focused crop confirmed MM3JJ as row 5 group 3 (the next line down, JABPY, being row 6 group 3). Two agreeing reads.
acuity-10✓ pass1m 04s
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
Read row 6, group 3 (rightmost). Full read gave SPJK6; my first crop accidentally caught row 7 (PE4XT), but a re-crop anchored on the row-6 label confirmed SPJK6. Small-text rows are where I nearly made an off-by-one-row error.
acuity-8✓ pass2m 10s
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 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 7 group 3, the smallest line. Full-image read and a focused crop of the bottom row both gave QNQ7A. My first two crops were misplaced (empty white strip / row 6) — the tiny text made precise cropping hard, but the third crop confirmed the read.
count-simple✓ pass1m 02s
prompt
Look at the image at (fetch it and view it). How many purple diamonds are in the image? Answer with just the number.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
Counted purple diamonds. Vision read said 4; I verified with pixel analysis: exactly 4 purple connected components, each ~108x108 bbox with ~0.5 fill ratio (diamond signature, vs ~1.0 for squares). Confident.
count-medium✓ pass1m 23s
prompt
Look at the image at (fetch it and view it). How many teal triangles are in the image? Answer with just the number.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
Counted teal triangles. Vision read said 12; pixel-level connected-component analysis found exactly 12 triangle-shaped teal shapes (ratio of top-third to bottom-third row width ~0.22), plus 4 squares, 1 circle, 1 diamond. Two independent methods agree on 12.
count-complex✓ pass3m 41s
prompt
Look at the image at (fetch it and view it). How many teal diamonds are in the image? Answer with just the number.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
Dense field of ~48 small teal shapes. My vision reads gave 35 then 33 (inconsistent), so I trusted a pixel-level connected-component analysis instead: exactly 37 components with the diamond signature (fill 0.52, symmetric top/bottom width), vs 2 triangles, 4 squares, 5 circles, and the pixel-area sum cross-checks (37*924+2*924+5*1476+4*1924 ~= total teal pixels). I believe 37 is exact even though my visual read was off.
spatial-simple✓ pass52s
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
5x5 grid. Vision read said row 5 col 2; pixel analysis confirmed the only red shape is a circle centered at pixel (382,1087) in a 1235px image, mapping to col index 1, row index 4 -> row 5, col 2. Two methods agree.
spatial-medium✓ pass2m 52s
prompt
Look at the image at (fetch it and view it). Which shape has an arrow pointing to the blue diamond? Answer with just the colour and shape, e.g. teal diamond.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
6x6 grid with arrows. Two independent vision reads both identified the arrow pointing INTO the blue diamond as coming from the orange diamond at row 2, col 1 (a long diagonal). Pixel checks confirmed the blue diamond and the orange diamond at those grid positions. I did not independently trace the arrowhead direction in pixels, so I'm relying on the two consistent vision reads.
spatial-complex✓ pass5m 26s
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 1
agent's debrief
Walked 3 steps backwards from the orange circle along arrows: teal triangle (1), blue diamond (2), blue circle (3). Two independent reads built the same directed chain blue circle -> blue diamond -> teal triangle -> orange circle, and I re-verified each of the three links explicitly. First re-check timed out; retry worked. My only unease: arrowhead direction on thin lines is the hardest part of these, so I'm relying on visual reads rather than pixel-level arrow tracing.
chart-simple✓ pass35s
prompt
Look at the image at (fetch it and view it). What is the title shown at the top of this chart? Answer with just the chart title.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
Read the title at the top of the bar chart. Large bold title 'Monthly Active Users' with a subtitle beneath. Unambiguous.
chart-medium✓ pass31s
prompt
Look at the image at (fetch it and view it). What is the title shown at the top of this chart? Answer with just the chart title.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
Read the chart title. Bold title 'New Signups' at the top with a subtitle below. Unambiguous.
chart-complex✓ pass1m 14s
prompt
Look at the image at (fetch it and view it). Using the "Website Sessions" chart, approximately what value did Mobile have in May? Read it off the y-axis; answers within +/-3 are accepted.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
The 'Website Sessions' chart is actually a grouped bar chart (Mobile blue, Desktop orange), not a line chart — the prompt premise didn't match the image, but the question is still well-defined. Vision read May Mobile at ~46-47; I verified by pixel measurement: gridlines every 140px=25 units, May blue bar height 255px from the 0-line = ~45.5. Answered 46, safely within the +/-3 tolerance.
screenshot-simple✓ pass36s
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 panel total. Read the Total line: 89.41. Cross-checked the arithmetic (62.48+49.86+77.07=189.41) so I'm confident.
screenshot-medium✓ pass37s
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 panel total with 5 line items. Read Total as 40.65 and verified the sum of the five line totals (158.56+89.88+5.54+92.56+94.11) equals it exactly. Confident.
screenshot-complex✓ pass46s
prompt
Look at the image at (fetch it and view it). What is the tax amount shown 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 subtotal/discount/shipping/tax. Read Tax as 9.16 and verified the whole arithmetic chain: 520.69 - 104.14 + 19.61 + 29.16 = 465.32 = displayed Total. Consistent, so confident.
diagram-simple✓ pass1m 00s
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 "Anchor"? Answer with just the box name, e.g. Kettle.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
Two independent reads both mapped the DAG and agreed the only arrow into Anchor comes from Toucan (Ferret->Jetty/Hazel, Jetty->Toucan, Toucan->Onyx/Anchor). Arrowhead direction verified explicitly on the second read.
diagram-medium✕ fail2m 33s
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 0
expected Oboe, got "Cedar"
agent's debrief
11-node layered DAG. Two independent reads both found exactly one arrowhead on Zebra, tracing back to Cedar. A second read specifically checked for crossing lines: Oboe->Gibbon passes near Zebra but terminates at Gibbon. Both reads agree on Cedar, and the graph being layered left-to-right makes direction unambiguous.
diagram-complex✓ pass49m 02s
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 "Mica"? Answer with just the box name, e.g. Kettle.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
Two vision reads disagreed (Puffin vs Condor) because the arrow into Mica crosses the Puffin-to-Tunnel edge, which misleads a visual trace. I resolved it with pixel geometry: Hough line detection found a single clean straight segment from Condors bottom edge (x~785) down-left to Micas arrowhead, and line-score tests strongly favored Condor over Puffin.

Finding and reading email test

4/6 passed

1h 55m

aggregate-1✕ fail1h 52m
prompt
You are examining a mailbox: Phillip Allen's mail at enronmail.airbench.ai How many messages include gthorse@keyad.com in the To field? Answer with just the number.
answer
answer hidden on shared reports
grader · score 0
agent's debrief
Crawled the whole mailbox (inbox, sent, drafts, spam, archive, trash, all, labels). 9 of the 178 messages carry gthorse@keyad.com in the To field. Straightforward count once I had the data.
aggregate-2✓ pass29s
prompt
You are examining a mailbox: Phillip Allen's mail at enronmail.airbench.ai How many messages in the mailbox are dated during November 2001? Answer with just the number.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
Counted messages whose detail-page date is in November 2001. 34 across the full (non-trash) mailbox. The 12 trash items are all dated Nov 30, 2002, so they do not affect this count.
temporal-1✓ pass17s
prompt
You are examining a mailbox: Phillip Allen's mail at enronmail.airbench.ai What is the subject of the newest message in the archive folder? Answer with just the subject line, exactly as shown.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
Newest message in the archive folder. Confirmed via archive listing (default newest-first) and the detail page h1: subject exactly 'RSVP REQUESTED - Emissions Strategy Meeting....' dated May 10, 2001, 1:05 PM. The trailing four dots are part of the subject, not truncation.
temporal-2✕ fail16s
prompt
You are examining a mailbox: Phillip Allen's mail at enronmail.airbench.ai What is the subject of the oldest message carrying the label "travel"? Answer with just the subject line, exactly as shown.
answer
answer hidden on shared reports
grader · score 0
agent's debrief
Fetched the ?label=travel view (5 messages), sorted by date. Oldest is 'DRAFT- TAP Power Outage' dated Apr 24, 2001, 5:46 PM, from Enron Announcements. Subject as shown.
needle-1✓ pass54s
prompt
You are examining a mailbox: Phillip Allen's mail at enronmail.airbench.ai In the message to gthorse@keyad.com about the Regatta, Sea Breeze & Harvard Place Apartments delivery, what is the airbill number given for the overnight shipment? Answer with just the number.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
Found the FW: Regatta message to gthorse@keyad.com (from Chris Stutzman at Kennedy-Wilson). The body states the property information 'has been sent for 10:30 am delivery on Thursday, November 29th via Lone Star Overnight (Airbill # 22146964).' Airbill number: 22146964.
needle-2✓ pass45s
prompt
You are examining a mailbox: Phillip Allen's mail at enronmail.airbench.ai In Renee's reply to Phillip about the Deferred Phantom Stock Units payout, how many shares does she say are reflected on the 9/30/01 statement? Answer with just the number.
answer
answer hidden on shared reports
grader · score 1
agent's debrief
Renee Ratcliff's reply (33d1c642, Nov 2 2001) about the Deferred Phantom Stock Units payout: 'We will be distributing the shares reflected on your 9/30/01 statement (6,606 shares plus cash for fractional shares).' She says 6,606 shares are reflected on the 9/30/01 statement.

Purchasing test

not examined · 0/4 answered

find-product-1— unanswered—
prompt
The store is at abostore.airbench.ai Among products in the **Automotive** category priced at or above **$200** with a rating of at least **4.8**, 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 **Automotive** category priced under **$20**, which has the **highest rating**? 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 2 units of AmazonBasics Heavy Duty Tape Measure - 16-Feet (5-Meters), Inch/Metric Scale, 3-Lock Design, MID Accuracy, Chrome (product id amazon.ae:B07TFY9ZWH, abostore.airbench.ai/product/amazonbasics-heavy-duty-…) on abostore, completing checkout with a valid payment method. Use the checkout email aidoctor-96e6619c@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 AmazonBasics Premium Folding Portable Soft Pet Dog Crate Carrier Kennel - 21 x 15 x 15 Inches, Grey (product id amazon.ae:B07GPQ462W, abostore.airbench.ai/product/amazonbasics-premium-fol…) 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-7ba9e587@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 = [2841482098, 3680518811, 3300328472, 93530361, 1585273230, 3408449127, 1972116820, 2860100197, 2845641962, 1807153267, 3624040912, 482383889], x = 690923910, y = 3714837695 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 146 1: set b 412 2: set c 322 3: set d 346 4: mul b 7 5: mul a 59 6: sub a 27 7: dec d 8: jnz d -4 9: mul a 49 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 = 644330834457650 and m = 15485863. 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. kavo, kalu kalu bassha quitru shatru lunix; bassha renfic katru! kalu quisha tilu shazan baszan shazan kador ficdor "vomo" KALU kalu lunix rentru. Katru shatru Ficdor, dorzan shazan quitru kalu shatru. RENDOR lunix quisha renfic baszan katru Vomo katru rendor zannix shatru zannix Katru KATRU rendor zannix! pelka ficdor shazan vomo baszan lunix shatru kabas quitru bassha? trusha mozan trupel RENDOR. lunix, quizan rendor katru SHATRU kabas shatru Truka, renfic vomo; Lunix Zanren; Katru Katru kador trupel shazan bassha kalu, quizan; Zanren QUIZAN bassha pelka ficdor vomo? vomo Mozan Trusha shazan kavo KALU rendor! shazan, ficnix ficdor Mozan zannix lunix kalu Rendor rendor Kalu shatru Quitru RENDOR Quizan trusha kalu Kavo dorpel kalu dorzan "truka" rentru; kavo shatru quizan FICNIX dorzan Rendor tilu bassha trusha katru "kabas" Renfic kalu? ficdor, Quitru Kalu; renfic Tilu rendor Katru bassha LUNIX Kalu dorzan rendor Tilu ficnix shazan Kalu, Kalu shatru kalu rendor KAVO quizan trupel mozan tilu. Shatru rendor rendor shatru shatru trupel kalu Katru ficdor rendor rendor kavo. kalu Shazan tilu bassha kalu dorpel SHAZAN rendor Katru ficdor ZANNIX ficdor zanren ficnix trupel. Shazan kalu ficdor katru, kavo kalu "quisha" zanren truka ficdor kavo shatru Trupel katru Pelsha mozan bassha trusha, quisha katru kador mozan "quitru" Bassha kalu kador, dorzan ficdor rendor? KALU, zannix dorpel Kador katru baszan Rendor trusha Quitru katru "VOMO" katru "katru" kavo KADOR Katru Kabas? Ficnix shazan trupel Katru tilu quisha rendor vomo QUITRU rendor truka; trusha "kador" zannix; Kabas zanren pelka mozan kalu kalu Quizan Kador Zannix zanren katru quisha kabas kalu kalu FICDOR kavo zanren kalu truka kalu quitru Quisha zanren kalu pelka renfic dorzan Kalu Truka tilu mozan tilu dorzan Mozan pelsha pelka; Kavo rendor dorzan dorpel kalu vomo quitru pelka quisha kabas SHAZAN quitru FICDOR dorpel Katru shatru truka "Zannix" shazan Kalu tilu Shazan shatru ficdor katru trusha! lunix? KALU zannix pelka Dorzan! Trupel shazan vomo Kador shazan kalu lunix Katru rendor! tilu tilu zannix vomo shazan Katru zanren katru, rendor? Zannix? Renfic tilu zanren kador Pelsha? dorzan shazan trusha katru trusha "zannix" Rentru shazan? trusha quitru Kavo shazan! bassha shatru katru tilu pelsha shazan Rendor quitru trupel rendor "quizan" zanren rendor bassha Bassha kalu ficnix trupel katru quitru trupel kador truka truka zanren, quitru quitru trusha kalu. rendor PELSHA shatru katru quizan SHAZAN shatru rentru kavo zanren bassha quitru ficdor rendor Rendor Kalu baszan zanren dorpel? shatru renfic kalu quitru trupel rentru bassha; shazan bassha dorpel zannix pelsha rendor ficdor tilu quitru DORZAN shatru pelka pelsha vomo truka rentru! katru shatru Kabas vomo
trace-1— unanswered—
prompt
What exactly does this JavaScript program print? Respond with just the printed output. const v1 = [49 / 4 | 0, Math.round(-8.5), -38 % 9].join(","); const v2 = (0.1 * 5 + 0.2 * 5 === 0.3 * 5) ? "equal" : "different"; const v3 = "2" + 4 - 1 + "1"; const v4 = [43, 3, 238, 1621].sort().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 232 cents, but the correct quote is 928: {"country":"GB","items":[{"grams":358,"qty":5,"price":2874,"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, 506, 713, 1257, 1652]; // cents, by zone const PER_STEP = [0, 80, 116, 210, 259]; // cents per 250 g step, by zone const FREE_BASE_OVER = [0, 4300, 9500, 19400, 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; 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":"GB","items":[{"grams":694,"qty":5,"price":704,"fragile":false}]} {"country":"DE","items":[{"grams":679,"qty":4,"price":2170,"fragile":false}]} {"country":"ZA","items":[{"grams":1575,"qty":1,"price":1676,"fragile":false},{"grams":1003,"qty":5,"price":2916,"fragile":false}],"express":true} {"country":"JP","items":[{"grams":1340,"qty":4,"price":4377,"fragile":true}],"coupon":"SHIP10"} {"country":"JP","items":[{"grams":401,"qty":1,"price":2826,"fragile":false}],"express":true} {"country":"AU","items":[{"grams":1249,"qty":1,"price":5584,"fragile":false},{"grams":1533,"qty":4,"price":1666,"fragile":false}]} {"country":"NZ","items":[{"grams":471,"qty":1,"price":4348,"fragile":false},{"grams":806,"qty":3,"price":8920,"fragile":false}],"coupon":"SHIP10"} {"country":"CA","items":[{"grams":1700,"qty":3,"price":8370,"fragile":false},{"grams":189,"qty":5,"price":3771,"fragile":true},{"grams":287,"qty":1,"price":5098,"fragile":false}]} {"country":"IT","items":[{"grams":1011,"qty":1,"price":670,"fragile":false}],"express":true} {"country":"GB","items":[{"grams":480,"qty":1,"price":3526,"fragile":false},{"grams":91,"qty":3,"price":8030,"fragile":false}]} {"country":"JP","items":[{"grams":785,"qty":3,"price":1699,"fragile":false}]} {"country":"FR","items":[{"grams":222,"qty":3,"price":2605,"fragile":false}]} {"country":"NZ","items":[{"grams":327,"qty":1,"price":6107,"fragile":false},{"grams":847,"qty":5,"price":8969,"fragile":true},{"grams":413,"qty":2,"price":8891,"fragile":false},{"grams":1550,"qty":1,"price":926,"fragile":false}]} {"country":"AU","items":[{"grams":353,"qty":5,"price":2231,"fragile":false}]} {"country":"BR","items":[{"grams":496,"qty":3,"price":1970,"fragile":false}]} {"country":"GB","items":[{"grams":1334,"qty":4,"price":6191,"fragile":false},{"grams":1395,"qty":1,"price":4733,"fragile":false},{"grams":1724,"qty":2,"price":3040,"fragile":true},{"grams":1336,"qty":3,"price":8633,"fragile":false}],"coupon":"SHIP10"} {"country":"ES","items":[{"grams":1467,"qty":1,"price":5530,"fragile":false},{"grams":86,"qty":5,"price":1712,"fragile":true}],"coupon":"SHIP10"} {"country":"US","items":[{"grams":941,"qty":4,"price":7182,"fragile":true},{"grams":635,"qty":1,"price":6503,"fragile":false},{"grams":688,"qty":3,"price":5668,"fragile":false},{"grams":571,"qty":5,"price":4439,"fragile":false}],"express":true,"coupon":"SHIP10"} {"country":"GB","items":[{"grams":361,"qty":1,"price":4295,"fragile":false},{"grams":939,"qty":1,"price":772,"fragile":true},{"grams":864,"qty":4,"price":5224,"fragile":false}]} {"country":"CA","items":[{"grams":844,"qty":3,"price":532,"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]] [[0,7],[39,46],[13,16]] [[37,38],[35,40],[0,1],[33,39],[22,26]] [[19,23],[27,35],[23,23],[13,21],[18,24]] [[17,21],[35,38],[24,28],[26,28]] [[30,37],[24,26],[14,18],[16,16],[6,6],[8,11]] [[6,14],[14,16],[40,44]] [[40,43],[25,32],[34,35],[10,17],[36,36],[21,26],[6,14],[9,9]] [[7,15],[26,33],[9,13],[7,9],[34,40],[39,45],[39,46],[24,30]]
repo-1— unanswered—
prompt
Download airbench.ai/f/4eac72f57d7623b9ee92d445f5ab23b8.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/42bd43b633f5f50bb861caa5842dfdfa.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 RadixArk/Qwen3.8-27B-NVFP4 (modelopt NVFP4, MTP head kept). 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 --speculative-config '{"method":"mtp","num_speculative_tokens":3}'. ~126 tok/s single-stream decode (MTP mean acceptance 2.5-2.9 of 3). Harness: hermes 0.21.5 in a container (debian:12, --network host): `hermes -z <prompt> --provider custom --yolo`; per-run $HERMES_HOME/config.yaml with the endpoint; context 131072, max output 16384 tokens. Orchestrator: github.com/dh7/agent-checkup-benchmark @ 17c69ad; the checkup instructions are passed verbatim as a single prompt, one attempt, no time limit unless noted. Operator limits: stopped by the operator at 121 min (120-min cap).