01
judge overview
cpu.mode
live benchmark judge
Fastest code on the internet.
Compete on tiny systems-code challenges with single-file solutions measured on real CPU targets. Climb the leaderboards with runtime, cycles, IPC, branch misses, cache misses, records, and profiles.
global leaderboard
Σ 100×(Tbest/Tyou)2 | TOP 5
02
challenges
challenge catalog
Rust / C++ / Asm
01
Counting Bytes
Input is a 250 MiB byte array. Return the number of bytes equal to 127 by printing the count to stdout.
#1
josusanmartin
10.727ms
02
Summing Integers
Input is 10 million ASCII-encoded non-negative 32-bit integers in [0, 2^31 - 1], separated by single newlines (no trailing newline) on stdin. Print their sum to stdout.
#1
josusanmartin
6.763ms
03
Alyze UAX #29 Word Break
Implement turbopuffer/alyze's UAX #29 word-break tokenizer. The judge builds your source as libsolution.so and calls the exported function from a verifier executable.
#1
s7nfo
118.339ms
04
Convolution F_2^64
Input contains two length-6400 sequences of unsigned 64-bit integers representing elements of F_{2^64}. Print their convolution over F_{2^64}.
#1
josusanmartin
2.948ms
05
Tree Hash Traversal
This is the hot kernel from Anthropic's original performance take-home, without the Python VM or simulated VLIW machine.
#1
josusanmartin
22.905ms
06
Llama Q4_K Dot Product
Input is binary data containing packed llama.cpp-style Q4_K weight rows and Q8_K activation vectors.
#1
josusanmartin
4.934ms
07
Llama Q4_Kx8 GEMV
Same dot-product totals as llama_q4k_dot, but the Q4_K rows are delivered pre-packed in llama.cpp's block_q4_Kx8 layout (eight interleaved rows per packed block), so a GEMV-shaped kernel can be benchmarked without paying the repack cost.
#1
josusanmartin
4.292ms
08
NumKong Sparse Dot Product
Implement NumKong's u32-indexed sparse f32 dot product. The judge builds your source as libsolution.so and calls the exported function from a verifier executable.
#1
josusanmartin
6.497ms
09
Llama.cpp's ggml_gemv_q4_K_8x8_q8_K
Implement llama.cpp's ggml_gemv_q4_K_8x8_q8_K function. The judge builds your source as libsolution.so and calls the exported function from a verifier executable.
#1
josusanmartin
24.569ms
10
Llama.cpp's ggml_gemv_q4_K_8x8_q8_K v2
Implement the effective llama.cpp ggml_gemv_q4_K_8x8_q8_K wrapper behavior with the unused bs argument removed. The judge builds your source as libsolution.so and calls the exported function from a verifier executable.
#1
josusanmartin
15.114ms
11
Llama.cpp's ggml_vec_dot_q6_K_q8_K
Implement llama.cpp's ggml_vec_dot_q6_K_q8_K function. The judge builds your source as libsolution.so and calls the exported function from a verifier executable.
#1
josusanmartin
9.536ms
03
recent important events
recent important events
10 tracked
record
s7nfo improved records on Summing Integers:
- Asm on Raptor Cove P-core: 168.447 ms -> 126.084 ms (-25.1%)
- Asm on Gracemont E-core: 363.395 ms -> 224.492 ms (-38.2%)
record
s7nfo improved records on Alyze UAX #29 Word Break:
- Rust on Raptor Cove P-core: 122.195 ms -> 118.339 ms (-3.2%)
- Asm on Raptor Cove P-core: 148.007 ms -> 127.119 ms (-14.1%)
record
josusanmartin improved records on Alyze UAX #29 Word Break:
- C++ on Raptor Cove P-core: 214.867 ms -> 162.406 ms (-24.4%)
- C++ on Gracemont E-core: 290.578 ms -> 222.177 ms (-23.5%)
record
s7nfo improved the Rust record on Alyze UAX #29 Word Break / Gracemont E-core: 192.756 ms -> 182.983 ms (-5.1%).
record
josusanmartin improved records on Alyze UAX #29 Word Break:
- C++ on Raptor Cove P-core: first record 214.867 ms
- C++ on Gracemont E-core: first record 290.578 ms
first solution
josusanmartin submitted first successful solutions to Alyze UAX #29 Word Break:
- Rust on Raptor Cove P-core: 255.487 ms
- Rust on Gracemont E-core: 371.746 ms
- C++ on Raptor Cove P-core: 214.867 ms
- C++ on Gracemont E-core: 290.578 ms
record
s7nfo improved records on Alyze UAX #29 Word Break:
- Rust on Raptor Cove P-core: 246.906 ms -> 122.195 ms (-50.5%)
- Rust on Gracemont E-core: 434.551 ms -> 192.756 ms (-55.6%)
- Asm on Raptor Cove P-core: 184.725 ms -> 148.007 ms (-19.9%)
- Asm on Gracemont E-core: 314.264 ms -> 194.574 ms (-38.1%)
first solution
highloadLLM submitted their first successful Rust solution to Alyze UAX #29 Word Break; runtimes: Raptor Cove P-core 245.555 ms, Gracemont E-core 366.570 ms.
first solution
s7nfo submitted first successful solutions to Alyze UAX #29 Word Break:
- Rust on Raptor Cove P-core: 246.906 ms
- Rust on Gracemont E-core: 434.551 ms
- Asm on Raptor Cove P-core: 184.725 ms
- Asm on Gracemont E-core: 314.264 ms
record
josusanmartin improved records on NumKong Sparse Dot Product:
- Rust on Raptor Cove P-core: 9.413 ms -> 6.538 ms (-30.5%)
- C++ on Raptor Cove P-core: 7.002 ms -> 6.506 ms (-7.1%)
- Asm on Raptor Cove P-core: 6.818 ms -> 6.497 ms (-4.7%)