Model comparison
DeepSeek V4.1 Flash vs Phi-1.5
DeepSeek V4.1 Flash has enough public results to be ranked (#40); Phi-1.5 does not yet, so treat this comparison as directional.
Last verified . 1 shared benchmarks.
DeepSeek
MicrosoftSummary
- They share 1 benchmark with published results for both.
Side by side
| DeepSeek V4.1 Flash | Phi-1.5 | |
|---|---|---|
| Provider | DeepSeek | Microsoft |
| Noometry Index | 50.3 | — |
| Released | 2026-09-09 | 2023-09-11 |
| Weights | Open | Open |
| Context window | 1M | — |
| Max output | 393K | — |
| Input $ / M tokens | $0.15 | — |
| Output $ / M tokens | $0.60 | — |
| Results tracked | 25 | 7 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
DeepSeek V4.1 Flash: 52.8 (#29), Phi-1.5: —
| Benchmark | DeepSeek V4.1 Flash | Phi-1.5 |
|---|---|---|
| LMArena WebDev | 1620 | — |
| SciCode | 51.9% | — |
| LMArena Coding | 1511 | — |
| ALE-Bench | 1,092 | — |
Reasoning Not comparable
DeepSeek V4.1 Flash: 46.2 (#39), Phi-1.5: —
| Benchmark | DeepSeek V4.1 Flash | Phi-1.5 |
|---|---|---|
| Epoch Capabilities Index | 155.01 | 90.82 |
| NYT Connections (extended) | 89.6% | — |
| CritPt | 14.3% | — |
| LMArena Hard Prompts | 1485 | — |
| DTBench | 89.9% | — |
| LMCA | 47% | — |
| Surface Evolver Bench | 46.3% | — |
| HellaSwag | — | 47.6% |
| WinoGrande | — | 73.4% |
Math Not comparable
DeepSeek V4.1 Flash: 48.4 (#52), Phi-1.5: —
| Benchmark | DeepSeek V4.1 Flash | Phi-1.5 |
|---|---|---|
| ProofBench | 54% | — |
| LMArena Math | 1493 | — |
Knowledge Not comparable
DeepSeek V4.1 Flash: 41.6 (#97), Phi-1.5: —
| Benchmark | DeepSeek V4.1 Flash | Phi-1.5 |
|---|---|---|
| LMArena Expert | 1505 | — |
| ARC (AI2) Challenge | — | 44.4% |
| BoolQ | — | 75.8% |
| MMLU | — | 37.6% |
| OpenBookQA | — | 37.2% |
Multilingual Not comparable
DeepSeek V4.1 Flash: 55.5 (#29), Phi-1.5: —
| Benchmark | DeepSeek V4.1 Flash | Phi-1.5 |
|---|---|---|
| LMArena Non-English | 1452 | — |
| LMArena Chinese | 1506 | — |
| LMArena French | 1448 | — |
| LMArena Russian | 1473 | — |
| LMArena Spanish | 1499 | — |
Instruction Following Not comparable
DeepSeek V4.1 Flash: 77.6 (#24), Phi-1.5: —
| Benchmark | DeepSeek V4.1 Flash | Phi-1.5 |
|---|---|---|
| LMArena Instruction Following | 1475 | — |
Long Context Not comparable
DeepSeek V4.1 Flash: 45.2 (#46), Phi-1.5: —
| Benchmark | DeepSeek V4.1 Flash | Phi-1.5 |
|---|---|---|
| LMArena Longer Query | 1476 | — |
Writing & Preference Not comparable
DeepSeek V4.1 Flash: 65.8 (#43), Phi-1.5: —
| Benchmark | DeepSeek V4.1 Flash | Phi-1.5 |
|---|---|---|
| LMArena Text | 1465 | — |
| LMArena Creative Writing | 1436 | — |
| EQ-Bench Creative Writing | 1540 | — |
| LMArena Multi-Turn | 1460 | — |
Frequently asked questions
Is DeepSeek V4.1 Flash better than Phi-1.5?
DeepSeek V4.1 Flash has enough public results to be ranked (#40); Phi-1.5 does not yet, so treat this comparison as directional.
How many benchmarks do DeepSeek V4.1 Flash and Phi-1.5 share?
1 benchmark has published results for both models. DeepSeek V4.1 Flash has 25 scored results on Noometry and Phi-1.5 has 7.