Agentic
Not comparable- LongCat-Flash-Lite-Sparse
- Not ranked
- Qwen3.5 397B
- 46.8
- Estimated · #71/152
- Basis
- BenchAlign lane · 4 vs 13 public rows
- Reading
- Not comparable
LongCat-Flash-Lite-Sparse vs Qwen3.5 397B
Updated September 10, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.
Decision reading
At least one model is not scored in the current public ranking lane, so the page does not name an overall quality winner.
10 results are shared. Category rows resting on Estimated evidence or different benchmark sets are marked directional and do not name a winner.
Share or export
Which one for your work
Recommendations appear only when a shared evidence basis or an explicit operating constraint supports the call. Secondary and unsupported use cases stay disclosed below the initial list.
Long documents
Prompts that approach the documented context limit
LongCat-Flash-Lite-Sparse
LongCat-Flash-Lite-Sparse has the larger documented context window.
Confidence: documented
Show secondary and unsupported calls
Coding work
Code generation, repair, and software-engineering tasks
Not enough matched evidence
LongCat-Flash-Lite-Sparse is not ranked on the public lane for coding, so no winner is named for coding.
Confidence: limited
Agentic work
Tool use, computer use, and multi-step task completion
Not enough matched evidence
LongCat-Flash-Lite-Sparse is not ranked on the public lane for agentic, so no winner is named for agentic.
Confidence: limited
Chat turn cost
1K fresh input + 500 output tokens
Not enough matched evidence
A complete comparable API-rate estimate is not available for both models.
Confidence: listed-rates
Cache-heavy agent loop cost
200K cached + 20K fresh input + 10K output tokens
Not enough matched evidence
The page does not recommend a cost winner because at least one model cannot fit the stated workload in one request. Qwen3.5 397B does not fit this workload in one request. Qwen3.5 397B has no published cached-input rate, so cached tokens use its listed input rate. LongCat-Flash-Lite-Sparse has no comparable published API token rate.
Confidence: rate-fallback
Repository review cost
50K fresh input + 3K output tokens
Not enough matched evidence
A complete comparable API-rate estimate is not available for both models.
Confidence: listed-rates
What is actually comparable
Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.
7 LongCat-Flash-Lite-Sparse only
10 Shared
28 Qwen3.5 397B only
- Shared results
- 10
- LongCat-Flash-Lite-Sparse only
- 7
- Qwen3.5 397B only
- 28
- Like-for-like categories
- 0 / 8
Category results, on a stated basis
Each row shows the public-lane category score for both models: the BenchAlign lane for agentic, coding, and knowledge, and the weighted public rows elsewhere, with the same rank each model page reports. A row is like-for-like only when both scores rest on Supported evidence or the same weighted set. Directional and not-comparable rows remain visible, but they never receive a winner.
Coding
Not comparable- LongCat-Flash-Lite-Sparse
- Not ranked
- Qwen3.5 397B
- 50.2
- Estimated · #59/151
- Basis
- BenchAlign lane · 4 vs 3 public rows
- Reading
- Not comparable
Reasoning
Not comparable- LongCat-Flash-Lite-Sparse
- Not ranked
- Qwen3.5 397B
- 59.8
- Unranked · 2 rankable rows
- Basis
- Provisional lane · 0 vs 1 weighted rows
- Reading
- Not comparable
Knowledge
Not comparable- LongCat-Flash-Lite-Sparse
- Not ranked
- Qwen3.5 397B
- 50.2
- Estimated · #77/183
- Basis
- BenchAlign lane · 5 vs 6 public rows
- Reading
- Not comparable
Math
Not comparable- LongCat-Flash-Lite-Sparse
- Not ranked
- Qwen3.5 397B
- 74.2
- Unranked · 5 rankable rows
- Basis
- Provisional lane · 2 vs 2 weighted rows
- Reading
- Not comparable
Multilingual
Not comparable- LongCat-Flash-Lite-Sparse
- Not ranked
- Qwen3.5 397B
- 69.7
- #5/12
- Basis
- Provisional lane · 0 vs 1 weighted rows
- Reading
- Not comparable
Multimodal
Not comparable- LongCat-Flash-Lite-Sparse
- Not ranked
- Qwen3.5 397B
- 63.0
- #27/48
- Basis
- Provisional lane · 0 vs 2 weighted rows
- Reading
- Not comparable
Instruction following
Not comparable- LongCat-Flash-Lite-Sparse
- Not ranked
- Qwen3.5 397B
- 0.0
- #123/123
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
| Category | LongCat-Flash-Lite-Sparse | Qwen3.5 397B | Basis | Reading |
|---|---|---|---|---|
| Agentic | Not ranked | 46.8Estimated · #71/152 | Not comparableBenchAlign lane · 4 vs 13 public rows | Not comparable |
| Coding | Not ranked | 50.2Estimated · #59/151 | Not comparableBenchAlign lane · 4 vs 3 public rows | Not comparable |
| Reasoning | Not ranked | 59.8Unranked · 2 rankable rows | Not comparableProvisional lane · 0 vs 1 weighted rows | Not comparable |
| Knowledge | Not ranked | 50.2Estimated · #77/183 | Not comparableBenchAlign lane · 5 vs 6 public rows | Not comparable |
| Math | Not ranked | 74.2Unranked · 5 rankable rows | Not comparableProvisional lane · 2 vs 2 weighted rows | Not comparable |
| Multilingual | Not ranked | 69.7#5/12 | Not comparableProvisional lane · 0 vs 1 weighted rows | Not comparable |
| Multimodal | Not ranked | 63.0#27/48 | Not comparableProvisional lane · 0 vs 2 weighted rows | Not comparable |
| Instruction following | Not ranked | 0.0#123/123 | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
Ranks count the models scored in each category’s lane, so the agentic, coding, and knowledge denominators (BenchAlign) differ from the provisional-lane categories. Unranked scores sit on the lane’s scale but never name a winner. Like-for-like rows within 0.5 points read as a practical tie.
Shape of the matched evidence
Only shared public evidence is shown. Sparse evidence stays a ruled list rather than being closed into a radar shape.
Too few matched category axes support a radar. The ruled list below shows only shared benchmark results; positions use each benchmark’s normalized display scale when available.
HMMT Feb 2026
Math
LongCat-Flash-Lite-Sparse: 40.5%Qwen3.5 397B: 87.9%Normalized gap 47.4LongCat-Flash-Lite-Sparse source Qwen3.5 397B sourceAIME26
Math
LongCat-Flash-Lite-Sparse: 65.7%Qwen3.5 397B: 93.3%Normalized gap 27.6LongCat-Flash-Lite-Sparse source Qwen3.5 397B sourceTerminal-Bench 2.0
Agentic
LongCat-Flash-Lite-Sparse: 33.7%Qwen3.5 397B: 52.5%Normalized gap 18.8LongCat-Flash-Lite-Sparse source Qwen3.5 397B sourceBrowseComp
Agentic
LongCat-Flash-Lite-Sparse: 48.6%Qwen3.5 397B: 62%Normalized gap 13.4LongCat-Flash-Lite-Sparse source Qwen3.5 397B sourceSWE-bench Pro
Coding
LongCat-Flash-Lite-Sparse: 40.6%Qwen3.5 397B: 50.9%Normalized gap 10.3LongCat-Flash-Lite-Sparse source Qwen3.5 397B source
What each workload costs
Three fixed token mixes turn per-token rates into comparable decisions. Each scenario states context fit and whether cached input had to fall back to the published list-input rate.
Chat turn
1K fresh input + 500 output tokens
- LongCat-Flash-Lite-Sparse
- Self-hosted; infrastructure cost varies
- Fits in one request
- Qwen3.5 397B
- $0.0024
- Fits in one request
LongCat-Flash-Lite-Sparse has no comparable published API token rate.
Repository review
50K fresh input + 3K output tokens
- LongCat-Flash-Lite-Sparse
- Self-hosted; infrastructure cost varies
- Fits in one request
- Qwen3.5 397B
- $0.0408
- Fits in one request
LongCat-Flash-Lite-Sparse has no comparable published API token rate.
Cache-heavy agent loop
200K cached + 20K fresh input + 10K output tokens
- LongCat-Flash-Lite-Sparse
- Self-hosted; infrastructure cost varies
- Fits in one request
- Cached-input rate unavailable
- Qwen3.5 397B
- $0.168
- Does not fit in one request
- Cached input priced at the published list-input rate
Qwen3.5 397B does not fit this workload in one request. Qwen3.5 397B has no published cached-input rate, so cached tokens use its listed input rate. LongCat-Flash-Lite-Sparse has no comparable published API token rate.
Specification differences
Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.
SpecificationLongCat-Flash-Lite-SparseQwen3.5 397B
Context window
Maximum documented context; output-token limits may be lower.
LongCat-Flash-Lite-Sparse
Qwen3.5 397B
128K
API model ID
LongCat-Flash-Lite-Sparse
meituan-longcat/LongCat-Flash-Lite-Sparse
Meituan LongCat-Flash-Lite-Sparse model cardQwen3.5 397B
Not sourced
Cached-input rate
A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.
LongCat-Flash-Lite-Sparse
No comparable hosted API rate
Meituan LongCat-Flash-Lite-Sparse model cardQwen3.5 397B
Not published
Documented inputs
LongCat-Flash-Lite-Sparse
Not sourced
Qwen3.5 397B
Not sourced
Documented outputs
LongCat-Flash-Lite-Sparse
Not sourced
Qwen3.5 397B
Not sourced
Provider availability
LongCat-Flash-Lite-Sparse
Not sourced
Qwen3.5 397B
Not sourced
Reasoning profile
LongCat-Flash-Lite-Sparse
Reasoning
Qwen3.5 397B
Non-Reasoning
Weight access
LongCat-Flash-Lite-Sparse
Open Weight
Qwen3.5 397B
Open Weight
License
LongCat-Flash-Lite-Sparse
Open Weight
Qwen3.5 397B
Open Weight
Release date
LongCat-Flash-Lite-Sparse
2026-07-31
Qwen3.5 397B
2026-02-16
If you already use one of these models
- Deployment change
- The models list different providers, so authentication, endpoint behavior, limits, and feature support may change.
- Quality signal
- At least one model is not scored in the current public ranking lane, so the page does not name an overall quality winner.
- Workload cost
- A complete comparable API-rate estimate is not available for both models.
- Context tradeoff
- LongCat-Flash-Lite-Sparse has the larger documented window (1M).
Run the same representative tasks against both endpoints before changing production traffic.
Benchmark evidence
The full public result ledger is available for audit without forcing a wide desktop table onto a phone.
Browse raw public benchmark evidence45 rows
Agentic
Terminal-Bench 2.0
LongCat-Flash-Lite-Sparse33.7%SourceQwen3.5 397B52.5%SourceQwen3.5 397B leads this result
VITA-Bench
LongCat-Flash-Lite-Sparse21.7%SourceQwen3.5 397B43.7%SourceQwen3.5 397B leads this result
MCP Atlas
LongCat-Flash-Lite-Sparse45.6%SourceQwen3.5 397B46.1%SourceQwen3.5 397B leads this result
BrowseComp
LongCat-Flash-Lite-Sparse48.6%SourceQwen3.5 397B62%SourceQwen3.5 397B leads this result
Claw-Eval
LongCat-Flash-Lite-Sparse—Qwen3.5 397B56.8%SourceNot directly comparable
QwenClawBench
LongCat-Flash-Lite-Sparse—Qwen3.5 397B51.8%SourceNot directly comparable
τ³-bench results
LongCat-Flash-Lite-Sparse—Qwen3.5 397B68.4%SourceNot directly comparable
DeepPlanning
LongCat-Flash-Lite-Sparse—Qwen3.5 397B37.6%SourceNot directly comparable
Toolathlon
LongCat-Flash-Lite-Sparse—Qwen3.5 397B36.3%SourceNot directly comparable
MCP-Tasks
LongCat-Flash-Lite-Sparse—Qwen3.5 397B74.2%SourceNot directly comparable
WideResearch
LongCat-Flash-Lite-Sparse—Qwen3.5 397B74.0%SourceNot directly comparable
Gert Labs
LongCat-Flash-Lite-Sparse—Qwen3.5 397B46.76%SourceNot directly comparable
ResearchClawBench
LongCat-Flash-Lite-Sparse—Qwen3.5 397B14.2%SourceNot directly comparable
Coding
SWE-bench Verified
LongCat-Flash-Lite-Sparse68.2%SourceQwen3.5 397B76.2%SourceQwen3.5 397B leads this result
SWE-bench Pro
LongCat-Flash-Lite-Sparse40.6%SourceQwen3.5 397B50.9%SourceQwen3.5 397B leads this result
SWE Multilingual
LongCat-Flash-Lite-Sparse59.3%SourceQwen3.5 397B—Not directly comparable
Terminal-Bench 2.0
LongCat-Flash-Lite-Sparse33.7%SourceQwen3.5 397B—Not directly comparable
LiveCodeBench v6
LongCat-Flash-Lite-Sparse—Qwen3.5 397B83.6%SourceNot directly comparable
Reasoning
Knowledge
MMLU
LongCat-Flash-Lite-Sparse85.3%SourceQwen3.5 397B—Not directly comparable
MMLU-Pro
LongCat-Flash-Lite-Sparse79.2%SourceQwen3.5 397B87.8%SourceQwen3.5 397B leads this result
CMMLU
LongCat-Flash-Lite-Sparse84.3%SourceQwen3.5 397B—Not directly comparable
C-Eval
LongCat-Flash-Lite-Sparse85.8%SourceQwen3.5 397B93%SourceQwen3.5 397B leads this result
GPQA-D
LongCat-Flash-Lite-Sparse69.5%SourceQwen3.5 397B—Not directly comparable
GPQA
LongCat-Flash-Lite-Sparse—Qwen3.5 397B88.4%SourceNot directly comparable
SuperGPQA
LongCat-Flash-Lite-Sparse—Qwen3.5 397B70.4%SourceNot directly comparable
MMLU-Redux
LongCat-Flash-Lite-Sparse—Qwen3.5 397B94.9%SourceNot directly comparable
HLE
LongCat-Flash-Lite-Sparse—Qwen3.5 397B28.7%SourceNot directly comparable
Math
MATH-500
LongCat-Flash-Lite-Sparse95.8%SourceQwen3.5 397B—Not directly comparable
AIME26
LongCat-Flash-Lite-Sparse65.7%SourceQwen3.5 397B93.3%SourceQwen3.5 397B leads this result
HMMT Feb 2026
LongCat-Flash-Lite-Sparse40.5%SourceQwen3.5 397B87.9%SourceQwen3.5 397B leads this result
IMOAnswerBench
LongCat-Flash-Lite-Sparse49.4%SourceQwen3.5 397B—Not directly comparable
HMMT Feb 2025
LongCat-Flash-Lite-Sparse—Qwen3.5 397B94.8%SourceNot directly comparable
HMMT Nov 2025
LongCat-Flash-Lite-Sparse—Qwen3.5 397B92.7%SourceNot directly comparable
MMAnswerBench
LongCat-Flash-Lite-Sparse—Qwen3.5 397B80.9%SourceNot directly comparable
Multilingual
Multimodal
MMMU-Pro
LongCat-Flash-Lite-Sparse—Qwen3.5 397B79%SourceNot directly comparable
MathVision
LongCat-Flash-Lite-Sparse—Qwen3.5 397B88.6%SourceNot directly comparable
CharXiv
LongCat-Flash-Lite-Sparse—Qwen3.5 397B80.8%SourceNot directly comparable
VideoMMMU
LongCat-Flash-Lite-Sparse—Qwen3.5 397B84.7%SourceNot directly comparable
ScreenSpot Pro
LongCat-Flash-Lite-Sparse—Qwen3.5 397B65.6%SourceNot directly comparable
V*
LongCat-Flash-Lite-Sparse—Qwen3.5 397B95.8%SourceNot directly comparable
Instruction following
IFEval
LongCat-Flash-Lite-Sparse—Qwen3.5 397B92.6%SourceNot directly comparable
Frequently asked questions
Which is better, LongCat-Flash-Lite-Sparse or Qwen3.5 397B?
At least one model is not scored in the current public ranking lane, so the page does not name an overall quality winner. The page therefore keeps the decision tied to the specific documented workload.
Which is better for coding, LongCat-Flash-Lite-Sparse or Qwen3.5 397B?
LongCat-Flash-Lite-Sparse is not ranked on the public lane for coding, so no winner is named for coding.
Which is better for agentic tasks, LongCat-Flash-Lite-Sparse or Qwen3.5 397B?
LongCat-Flash-Lite-Sparse is not ranked on the public lane for agentic tasks, so no winner is named for agentic tasks.
Which costs less, LongCat-Flash-Lite-Sparse or Qwen3.5 397B?
Both models do not have comparable published API token rates, so this page does not name a universal price winner.
Which has the larger context window, LongCat-Flash-Lite-Sparse or Qwen3.5 397B?
LongCat-Flash-Lite-Sparse has the larger documented context window: 1M, compared with 128K.
Related comparisons
Last updated September 10, 2026
Watch LongCat-Flash-Lite-Sparse vs Qwen3.5 397B
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 2,000+ readers.