Meituan model product
LongCat Flash Lite
5B activated per token.
Updated Oct 23, 2026. Default version: LongCat-Flash-Lite
LLMBoard Score30.6LongCat-Flash-Lite
Coverage100%13 benchmark families
Context window256KTokens
Official input priceN/AOfficial price unavailable
LongCat Flash Lite Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
LongCat Flash Lite Benchmark Results
Benchmark scores for LongCat-Flash-Lite.
Sort by
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkCMMLU | Score82.48% | Rank04 | Participants6 | Percentile40.00% | EvidenceC | Evaluated |
| BenchmarkMATH-500 | Score96.80% | Rank10 | Participants33 | Percentile71.88% | EvidenceC | Evaluated |
| BenchmarkTerminal-Bench | Score33.75% | Rank15 | Participants25 | Percentile41.67% | EvidenceC | Evaluated |
| BenchmarkTau2 Airline | Score58.00% | Rank16 | Participants26 | Percentile40.00% | EvidenceC | Evaluated |
| BenchmarkTau2 Retail | Score73.10% | Rank17 | Participants28 | Percentile40.74% | EvidenceC | Evaluated |
| BenchmarkTau2 Telecom | Score72.80% | Rank27 | Participants37 | Percentile27.78% | EvidenceC | Evaluated |
| BenchmarkAIME 2024 | Score72.19% | Rank39 | Participants54 | Percentile28.30% | EvidenceC | Evaluated |
| BenchmarkMMLU | Score85.52% | Rank39 | Participants103 | Percentile62.75% | EvidenceC | Evaluated |
| BenchmarkSWE-bench Multilingual | Score38.10% | Rank48 | Participants50 | Percentile4.08% | EvidenceC | Evaluated |
| BenchmarkMMLU-Pro | Score78.29% | Rank67 | Participants142 | Percentile53.19% | EvidenceC | Evaluated |
| BenchmarkSWE-Bench Verified | Score54.40% | Rank99 | Participants120 | Percentile17.65% | EvidenceC | Evaluated |
| BenchmarkAA IFBench | Score43.06% | Rank100 | Participants168 | Percentile40.72% | EvidenceB | EvaluatedN/A |
| BenchmarkAIME 2025 | Score63.23% | Rank101 | Participants122 | Percentile17.36% | EvidenceC | Evaluated |
| BenchmarkAA LCR v1.1 | Score28.33% | Rank150 | Participants193 | Percentile22.40% | EvidenceB | EvaluatedN/A |
| BenchmarkAA GPQA Diamond | Score63.64% | Rank151 | Participants199 | Percentile24.24% | EvidenceB | EvaluatedN/A |
| BenchmarkAA HLE Text No Tools | Score5.84% | Rank152 | Participants200 | Percentile24.12% | EvidenceB | EvaluatedN/A |
| BenchmarkGPQA | Score66.78% | Rank159 | Participants256 | Percentile38.04% | EvidenceC | Evaluated |
| BenchmarkAA CritPt | Score0.00% | Rank165 | Participants200 | Percentile17.59% | EvidenceB | EvaluatedN/A |
| BenchmarkAA Omniscience Accuracy | Score13.30% | Rank175 | Participants201 | Percentile13.00% | EvidenceB | EvaluatedN/A |
LongCat Flash Lite Arena Results
Preference and agent-evaluation results for the default version.
No Arena results
The default version does not have a matching Arena result yet.
LongCat Flash Lite Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- N/A
- Official provider
- N/A
- Lowest third-party
- N/A
- Tracked offerings
- 0
No provider prices
The default version has no current input or output token prices.
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
LongCat Flash Lite Runtime Performance
Provider-specific output speed and catalog latency for LongCat-Flash-Lite. Runtime does not affect the capability score.
No runtime data
No provider-specific speed or latency record is linked to the default version yet.
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
LongCat Flash Lite Specifications
Technical details for the model's default version.
- Version
- LongCat-Flash-Lite
- Released
- Feb 5, 2026
- Knowledge cutoff
- Unknown
- Parameters
- 68.5B
- Context window
- 256K
- Max output
- 128K
- Inputs
- text
- Outputs
- text
- Open weights
- Yes
- License
- MIT
LongCat Flash Lite Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
Sort by
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionLongCat-Flash-Lite | Released | LLMBoard30.62 | Parameters68.5B | Context256K | Max output128K | Open weightsYes | LicenseMIT |
Models similar to LongCat Flash Lite
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
ME
LongCat Flash Chat
Meituan
ME
LongCat Flash Thinking
Meituan
MA
Mistral Small 4
Mistral AI
NV
Nemotron Nano 9B
NVIDIA
LA
LFM2.5 2.6B
Liquid AI
AC
Qwen3.5 4B
Alibaba Cloud / Qwen Team
What is LongCat Flash Lite?
Key information about LongCat Flash Lite and its available data.
LongCat-Flash-Lite is a Meituan MoE model that uses N-gram embedding expansion and supports a 256K context length via YaRN. It is designed for agent tooling and programming tasks, with reported inference speeds of 500–700 tokens per second.
Data as of 2026-10-22.
FAQ
Common questions about LongCat Flash Lite.
When was LongCat Flash Lite released?
LongCat Flash Lite's default version was released on Feb 5, 2026.
How much does LongCat Flash Lite cost?
No official standard PAYG price is currently available for LongCat Flash Lite.
Who created LongCat Flash Lite?
LongCat Flash Lite was created by Meituan.
What is the context window for LongCat Flash Lite?
The default version has a 256K token context window.
Is LongCat Flash Lite open weight?
Yes. The default version is marked as open weight under MIT.
How many API providers offer LongCat Flash Lite?
No provider offering is currently linked to the default version.