LongCat-Flash-Lite-Sparse

LongCat-Flash-Lite-Sparse is a reasoning model from Meituan in the LongCat-Flash-Lite family. 10 benchmarks count toward its score, in 4 categories.

availableShows if the model has enough results for an index.
IndexOverall score out of 100.41.5 ±8.6
CoverageShare of the index weight with results.65%
SpeedOutput tokens per second.
Input / 1MUS dollars per 1M input tokens.Free
Output / 1MUS dollars per 1M output tokens.Free
ContextMaximum tokens in one request.1M
EloLMArena rating and rank.N/A

The index is a score out of 100. The ± range shows how much it can change.

CapabilitiesScore per category, out of 100.

Out of 100
AgenticMulti-step tasks with tools.
41.4
CodingCode writing and repair.
47.8
ReasoningLogic problems and puzzles.
N/A
MultimodalTasks with images and text.
N/A
KnowledgeFacts and expert knowledge.
43.9
MultilingualTasks in many languages.
N/A
InstructionTasks with strict rules in the prompt.
N/A
MathMath problems.
33.8

Results

10 counted
BenchmarkThe test name.CategoryThe capability that the test measures.ResultThe score from the publisher.IndexThis result as a score out of 100.RunThe settings of the run.DateDate of the result.Published byThe source of the result.
MATH-500 Problem SetMath95.8%48.9Dan Hendrycks et al.
C-EvalKnowledge85.8%C-Eval authors
Massive Multitask Language UnderstandingKnowledge85.3%Dan Hendrycks et al.
Chinese Massive Multitask Language UnderstandingKnowledge84.3%DeepSeek-AI
Massive Multitask Language Understanding ProfessionalKnowledge79.2%45.4Yubo Wang et al.
GPQA DiamondKnowledge69.5%42.4David Rein et al.
Software Engineering Benchmark VerifiedCoding68.2%52.2Carlos E. Jimenez et al.
AIME 2026Math65.7%32.3Qwen
IMOAnswerBenchMath49.4%DeepSeek-AI
BrowseCompAgentic48.6%40.8OpenAI
MCP AtlasAgentic45.6%42.6OpenAI
SWE-bench ProCoding40.6%43.3Xiang Deng et al.
Harvard-MIT Mathematics Tournament February 2026Math40.5%20.2Qwen
VITA-BenchAgentic21.7%40.9Meituan LongCat Team

10 benchmarks count, from 10 of 14 results. A grey row does not count. Too few models took that benchmark.

Sources

BenchLM benchmark aggregationUsed with attribution; per-benchmark results credited to their original authors

More from Meituan

LongCat-2.0Unranked