Inference
EN
Meituan

LongCat-Flash-Lite-Sparse

LongCat-Flash-Lite-Sparse is a reasoning model from Meituan in the LongCat-Flash-Lite family. 14 benchmarks count toward its index, in 4 categories.

  • Open weights
  • Reasoning
  • Input: text
  • Released 2026-07
RankedEnough results for an index.
IndexAll categories, out of 100.39.3±7.8Median 51.0
CoverageIndex weight that has results.65%
SpeedOutput tokens per second.—Median 67/s
First tokenSeconds to the first output token.—Median 1.3 s
Input / 1MUS dollars per 1M input tokens.—Median $0.60
Output / 1MUS dollars per 1M output tokens.—Median $2.55
ContextMaximum tokens in one request.1MMedian 262K
EloLMArena rating and rank.—Median 1422

Run locally

69.13B parameters
16-bit
146.2 GB
8-bit
78.1 GB
4-bit
45.0 GB
Weights
Hugging Face (opens in a new tab) ↗

Memory need with a short context. The 4-bit copy fits in 48 GB. Compare local models

Capabilities

Index per category, out of 100.
Out of 100
AgenticMulti-step tasks with tools.4 counted39.6180 / 218
CodingCode writing and repair.2 counted47.5152 / 209
ReasoningLogic problems and puzzles.None counted——
MultimodalTasks with images and text.None counted——
KnowledgeFacts and expert knowledge.4 counted39.7195 / 268
MultilingualTasks in many languages.None counted——
InstructionTasks with strict prompt rules.None counted——
MathMath problems.4 counted32.5139 / 168

Results

14 counted
BenchmarkThe benchmark name.ResultThe score from the source.PlacePlace among results on this benchmark.
Massive Multitask Language UnderstandingKnowledge85.3%18 / 34
Massive Multitask Language Understanding ProfessionalKnowledge79.2%40 / 75
MATH-500 Problem SetMath95.8%16 / 27
Software Engineering Benchmark VerifiedCoding68.2%77 / 114
VITA-BenchAgentic21.7%10 / 13
GPQA Diamond · BenchLMKnowledge69.5%73 / 84
AIME 2026Math65.7%31 / 34
C-EvalKnowledge85.8%11 / 12
MCP AtlasAgentic45.6%48 / 52
SWE-bench ProCoding40.6%78 / 83
BrowseCompAgentic48.6%60 / 63
Harvard-MIT Mathematics Tournament February 2026Math40.5%26 / 27
IMOAnswerBenchMath49.4%10 / 10
Terminal-Bench 2.0 · BenchLMAgentic33.7%41 / 41
Chinese Massive Multitask Language UnderstandingKnowledge84.3%—

Sources

BenchLMCC BY-NC 4.0

History

5 releases
Index40.0 → 39.3
23 Oct 202612 Nov 2026

Questions

How good is LongCat-Flash-Lite-Sparse?
Its index is 39.3 out of 100. That is place 211 of 283 ranked models.
What is LongCat-Flash-Lite-Sparse best at?
Coding: 47.5 out of 100, place 152 of 209.
What is the context window of LongCat-Flash-Lite-Sparse?
1M tokens.
Is LongCat-Flash-Lite-Sparse open source?
Yes. Its weights are published.
When was LongCat-Flash-Lite-Sparse released?
July 2026.

Badge

LongCat-Flash-Lite-Sparse on Inference360: place 211 of 283, index 39.3
HTML
<a href="https://inference360.ai/models/longcat-flash-lite-sparse"><img src="https://inference360.ai/badges/longcat-flash-lite-sparse.svg" alt="LongCat-Flash-Lite-Sparse on Inference360: place 211 of 283, index 39.3" height="20"></a>
Markdown
[![LongCat-Flash-Lite-Sparse on Inference360: place 211 of 283, index 39.3](https://inference360.ai/badges/longcat-flash-lite-sparse.svg)](https://inference360.ai/models/longcat-flash-lite-sparse)

Compare with

Kimi K239.6

More from Meituan

All models
LongCat-2.0UnrankedLongCat-2.5Unranked