Model comparison
Longcat Flash Chat vs Phi-4
Longcat Flash Chat is the stronger model overall, scoring 41.5 to 30.9 on the Noometry Index.
Last verified . 17 shared benchmarks.
MicrosoftSummary
- They share 17 benchmarks with published results for both. Longcat Flash Chat scores higher in 8 categories and Phi-4 in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Longcat Flash Chat leads 61.2 to 40.7.
Side by side
| Longcat Flash Chat | Phi-4 | |
|---|---|---|
| Provider | Meituan | Microsoft |
| Noometry Index | 41.5 | 30.9 |
| Released | — | 2024-12-11 |
| Weights | Open | Open |
| Context window | — | 128K |
| Max output | — | 4K |
| Input $ / M tokens | — | $0.07 |
| Output $ / M tokens | — | $0.14 |
| Results tracked | 19 | 37 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Longcat Flash Chat leads
Longcat Flash Chat: 42.9 (#85), Phi-4: 33.8 (#237)
| Benchmark | Longcat Flash Chat | Phi-4 |
|---|---|---|
| LMArena Coding | 1471 | 1231 |
| BigCodeBench Instruct | — | 45.5% |
| LiveBench Coding | — | 30.7% |
| BigCodeBench Complete | — | 55.4% |
Agentic & Tool Use Not comparable
Longcat Flash Chat: —, Phi-4: 22.6 (#120)
| Benchmark | Longcat Flash Chat | Phi-4 |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 28.8% |
| BALROG | — | 11.6% |
Reasoning Longcat Flash Chat leads
Longcat Flash Chat: 18.3 (#267), Phi-4: 17.0 (#287)
| Benchmark | Longcat Flash Chat | Phi-4 |
|---|---|---|
| LMArena Hard Prompts | 1440 | 1220 |
| Kagi LLM Benchmark | 43.9% | — |
| NYT Connections (extended) | 17.7% | — |
| Chess Puzzles | — | 1% |
| LiveBench Reasoning | — | 47.8% |
| LiveBench Data Analysis | — | 45.2% |
| Epoch Capabilities Index | — | 130.43 |
| LiveBench | — | 41.6% |
Math Longcat Flash Chat leads
Longcat Flash Chat: 36.6 (#107), Phi-4: 19.6 (#279)
| Benchmark | Longcat Flash Chat | Phi-4 |
|---|---|---|
| LMArena Math | 1441 | 1246 |
| OTIS Mock AIME 2024-2025 | — | 13.8% |
| LiveBench Math | — | 42% |
| MATH Level 5 | — | 64.9% |
Knowledge Longcat Flash Chat leads
Longcat Flash Chat: 40.0 (#111), Phi-4: 32.5 (#204)
| Benchmark | Longcat Flash Chat | Phi-4 |
|---|---|---|
| LMArena Expert | 1454 | 1203 |
| GPQA Diamond | — | 56.1% |
| Confabulations | — | 29.4% |
| Vectara Hallucination Rate | — | 3.7% |
| MMLU | — | 84.8% |
Multilingual Longcat Flash Chat leads
Longcat Flash Chat: 52.1 (#95), Phi-4: 37.4 (#232)
| Benchmark | Longcat Flash Chat | Phi-4 |
|---|---|---|
| LMArena Non-English | 1404 | 1197 |
| LMArena Chinese | 1465 | 1211 |
| LMArena French | 1457 | 1223 |
| LMArena German | 1409 | 1222 |
| LMArena Japanese | 1373 | 1158 |
| LMArena Korean | 1372 | 1151 |
| LMArena Russian | 1396 | 1209 |
| LMArena Spanish | 1445 | 1234 |
Instruction Following Longcat Flash Chat leads
Longcat Flash Chat: 74.6 (#92), Phi-4: 60.6 (#246)
| Benchmark | Longcat Flash Chat | Phi-4 |
|---|---|---|
| LMArena Instruction Following | 1411 | 1201 |
| LiveBench Instruction Following | — | 58.4% |
Long Context Longcat Flash Chat leads
Longcat Flash Chat: 43.5 (#89), Phi-4: 36.9 (#221)
| Benchmark | Longcat Flash Chat | Phi-4 |
|---|---|---|
| LMArena Longer Query | 1425 | 1217 |
Writing & Preference Longcat Flash Chat leads
Longcat Flash Chat: 61.2 (#88), Phi-4: 40.7 (#239)
| Benchmark | Longcat Flash Chat | Phi-4 |
|---|---|---|
| LMArena Text | 1427 | 1217 |
| LMArena Creative Writing | 1388 | 1182 |
| LMArena Multi-Turn | 1418 | 1206 |
| Short-Story Creative Writing | — | 62.6% |
| LiveBench Language | — | 25.6% |
Frequently asked questions
Is Longcat Flash Chat better than Phi-4?
Longcat Flash Chat is the stronger model overall, scoring 41.5 to 30.9 on the Noometry Index.
Is Longcat Flash Chat or Phi-4 better for coding?
Longcat Flash Chat scores higher on coding benchmarks: 42.9 versus 33.8 in the Noometry coding category.
How many benchmarks do Longcat Flash Chat and Phi-4 share?
17 benchmarks have published results for both models. Longcat Flash Chat has 19 scored results on Noometry and Phi-4 has 37.