Model comparison
DeepSeek V4 Flash vs Phi-3.5-mini
DeepSeek V4 Flash has enough public results to be ranked (#35); Phi-3.5-mini does not yet, so treat this comparison as directional.
Last verified . 0 shared benchmarks.
DeepSeek
MicrosoftSummary
- The widest gap is in coding, where DeepSeek V4 Flash leads 47.9 to 34.6.
Side by side
| DeepSeek V4 Flash | Phi-3.5-mini | |
|---|---|---|
| Provider | DeepSeek | Microsoft |
| Noometry Index | 53.6 | 36.9 |
| Released | 2026-04-24 | 2024-08-16 |
| Weights | Open | Open |
| Context window | 1M | — |
| Max output | 393K | — |
| Input $ / M tokens | $0.15 | — |
| Output $ / M tokens | $0.60 | — |
| Results tracked | 41 | 5 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek V4 Flash leads
DeepSeek V4 Flash: 47.9 (#59), Phi-3.5-mini: 34.6 (#232)
| Benchmark | DeepSeek V4 Flash | Phi-3.5-mini |
|---|---|---|
| FrontierCode | 18.8% | — |
| LMArena WebDev | 1582 | — |
| SciCode | 49.9% | — |
| WeirdML | 63% | — |
| BigCodeBench Instruct | — | 32.8% |
| LMArena Coding | 1457 | — |
| BigCodeBench Complete | — | 38.5% |
| ALE-Bench | 1,306 | — |
Reasoning Not comparable
DeepSeek V4 Flash: 53.7 (#30), Phi-3.5-mini: —
| Benchmark | DeepSeek V4 Flash | Phi-3.5-mini |
|---|---|---|
| ARC-AGI-2 | 61.4% | — |
| SimpleBench | 61.1% | — |
| Kagi LLM Benchmark | 52.2% | — |
| NYT Connections (extended) | 89.6% | — |
| ARC-AGI-1 | 89% | — |
| CritPt | 16.6% | — |
| Chess Puzzles | 33% | — |
| LMArena Hard Prompts | 1444 | — |
| Mystery Game Puzzles | 34% | — |
| DTBench | 90.9% | — |
| LMCA | 41.7% | — |
| Epoch Capabilities Index | 154.49 | — |
| PIQA | — | 81% |
Math Not comparable
DeepSeek V4 Flash: 60.3 (#37), Phi-3.5-mini: —
| Benchmark | DeepSeek V4 Flash | Phi-3.5-mini |
|---|---|---|
| FrontierMath (Tiers 1-3) | 57.5% | — |
| FrontierMath Tier 4 | 24.4% | — |
| MathArena Final-Answer Competitions | 76.5% | — |
| OTIS Mock AIME 2024-2025 | 94.4% | — |
| ProofBench | 56% | — |
| LMArena Math | 1427 | — |
| GSM8K | — | 86.2% |
Knowledge Not comparable
DeepSeek V4 Flash: 55.4 (#48), Phi-3.5-mini: —
| Benchmark | DeepSeek V4 Flash | Phi-3.5-mini |
|---|---|---|
| GPQA Diamond | 91% | — |
| SimpleQA Verified | 33.6% | — |
| LMArena Expert | 1441 | — |
| BoolQ | — | 78% |
Multilingual Not comparable
DeepSeek V4 Flash: 53.0 (#72), Phi-3.5-mini: —
| Benchmark | DeepSeek V4 Flash | Phi-3.5-mini |
|---|---|---|
| LMArena Non-English | 1420 | — |
| LMArena Chinese | 1468 | — |
| LMArena French | 1439 | — |
| LMArena German | 1418 | — |
| LMArena Japanese | 1406 | — |
| LMArena Korean | 1384 | — |
| LMArena Russian | 1428 | — |
| LMArena Spanish | 1436 | — |
Instruction Following Not comparable
DeepSeek V4 Flash: 74.9 (#81), Phi-3.5-mini: —
| Benchmark | DeepSeek V4 Flash | Phi-3.5-mini |
|---|---|---|
| LMArena Instruction Following | 1421 | — |
Long Context Not comparable
DeepSeek V4 Flash: 43.8 (#85), Phi-3.5-mini: —
| Benchmark | DeepSeek V4 Flash | Phi-3.5-mini |
|---|---|---|
| LMArena Longer Query | 1434 | — |
Writing & Preference Not comparable
DeepSeek V4 Flash: 63.8 (#61), Phi-3.5-mini: —
| Benchmark | DeepSeek V4 Flash | Phi-3.5-mini |
|---|---|---|
| LMArena Text | 1432 | — |
| LMArena Creative Writing | 1403 | — |
| EQ-Bench Creative Writing | 1559 | — |
| LMArena Multi-Turn | 1449 | — |
Frequently asked questions
Is DeepSeek V4 Flash better than Phi-3.5-mini?
DeepSeek V4 Flash has enough public results to be ranked (#35); Phi-3.5-mini does not yet, so treat this comparison as directional.
Is DeepSeek V4 Flash or Phi-3.5-mini better for coding?
DeepSeek V4 Flash scores higher on coding benchmarks: 47.9 versus 34.6 in the Noometry coding category.
How many benchmarks do DeepSeek V4 Flash and Phi-3.5-mini share?
0 benchmarks have published results for both models. DeepSeek V4 Flash has 41 scored results on Noometry and Phi-3.5-mini has 5.