Passedleetcode or numeric
Task ID 2808
Completed the recorded checks.
Exact task metadata
2808 · Task ID 2808
No recorded stop reason
Reward1
Assertions—
Steps—
Tool calls—
Model calls—
Tokens in / out467 / 2,965
CodingCode generation
LiveCodeBench v6
Version 6.0 · 93 tasks. Exact task sets and protocols remain separate.
Can it turn a problem into working code?
90.3%Recorded score · 84 / 93 tasks
DeepSeek V4 Flash · UD-IQ3_XXS · GB10 · 128GB
Inspect the measured run ↗How scoring works ↓THE QUESTION IT ANSWERS
Can it turn a problem into working code?
Useful for understanding self-contained coding ability, where correctness can be checked by running the answer.
- 01Programming problem→
- 02Generate a solution→
- 03Executable tests
Interpret the score Version and problem window matter. This is different from maintaining an existing codebase.
BEYOND THE HEADLINE
What happened across 100 tasks?
Each block represents one recorded task, in receipt order from left to right, then top to bottom. Fill shows the outcome; work-category colors are not used here.
Fully passed84 84.0%
No credit16 16.0%
Tested by LocalsOnly · UD-IQ3_XXS · GB10 · 128GB. Partial progress still requires further work before a task is complete.
Inspect individual task outcomes →What this test is
What this benchmark measures
Solve programming problems with executable tests.
Time-indexed programming problems support generation, repair, execution and prediction tests. Version, time window and scenario matter when interpreting a result.
LiveCodeBench paper ↗How it grades
How a task earns its score
The LocalsOnly measured receipt records each task as pass or fail.
A task can also carry a setup issue recorded on the receipt (for example checker_fail).
0 infrastructure outcomes are recorded on the receipt. They are not invented extra grades.
Published rows are vendor or bench cites. They do not share the measured task grid.
SuiteLiveCodeBench v6
Tasks100 on the measured receipt
PassRecorded pass on the task
FailRecorded fail on the task
Setup issuechecker_fail
Follow one task · this measured recipe
A passed task and a failed one
Same LocalsOnly receipt on GB10 · 128GB. DeepSeek V4 Flash · UD-IQ3_XXS · 2026-09-12. Both from leetcode or numeric. Assertions, steps, and tool calls are on the task record. There is no agent transcript on this page.
passed vs checker_fail
Not passedleetcode or numeric
Task ID 3308
Did not complete the recorded checks. The record does not establish a cause.
Exact task metadata
3308 · Task ID 3308
checker_fail · Official LCB checker returned fail on nonempty generation
Reward0
Assertions—
Steps—
Tool calls—
Model calls—
Tokens in / out351 / 4,348
Subject & category breakdown · this recipe
Results by subject or work category
Ranked by this file’s recorded pass rate. A blank rate is not a zero. Names come from the receipt.
- codeforces like100.0%1/1
- leetcode or numeric95.6%43/45
- atcoder abc88.4%38/43
- atcoder arc50.0%2/4
Ranking
Results & published references
Place is assigned only among rows that share a comparable score on this board. Evaluation settings can differ across sources.
Tested by LocalsOnly
Results from the exact local setup shown below.
| # | Model / recipe | Reported score ↓ | Runs on | Evidence | |
|---|---|---|---|---|---|
| 01 | ◎DeepSeek V4 Flash · llama.cppUD-IQ3_XXS · llama.cpp | 90.3% | GB10 · 128GB | ● Tested by LocalsOnlyLocalsOnly | ↗ |
| 02 | ✳Qwen 3.8 27B · llama.cppUD-Q4_K_XL · llama.cpp | 66.0% | GB10 · 128GB | ● Tested by LocalsOnlyLocalsOnly | ↗ |
Published reference
Results reported by model developers and benchmark authors.
| # | Model / recipe | Reported score ↓ | Runs on | Evidence | |
|---|---|---|---|---|---|
| — | ✳Qwen 3.8 27BBF16 reference · Reported evaluation | 90.3% | Published reference | ↗ Published resultQwen | ↗ |