Model comparison
GPT-6 Sol vs Phi-3.5-mini
GPT-6 Sol has enough public results to be ranked (#46); Phi-3.5-mini does not yet, so treat this comparison as directional.
Last verified . 0 shared benchmarks.
MicrosoftSummary
- The widest gap is in coding, where GPT-6 Sol leads 59.6 to 34.0.
- Phi-3.5-mini has downloadable open weights; the other is API-only.
Side by side
| GPT-6 Sol | Phi-3.5-mini | |
|---|---|---|
| Provider | OpenAI | Microsoft |
| Noometry Index | 49.2 | 36.4 |
| Released | 2026-09-22 | 2024-08-16 |
| Weights | Proprietary | Open |
| Context window | 1.05M | — |
| Max output | 128K | — |
| Input $ / M tokens | $2 | — |
| Output $ / M tokens | $10 | — |
| Results tracked | 22 | 5 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-6 Sol leads
GPT-6 Sol: 59.6 (#11), Phi-3.5-mini: 34.0 (#231)
| Benchmark | GPT-6 Sol | Phi-3.5-mini |
|---|---|---|
| DeepSWE | 68.8% | — |
| FrontierCode | 49.3% | — |
| LMArena WebDev | 1692 | — |
| BigCodeBench Instruct | — | 32.8% |
| LMArena Coding | 1453 | — |
| BigCodeBench Complete | — | 38.5% |
Agentic & Tool Use Not comparable
GPT-6 Sol: 33.5 (#49), Phi-3.5-mini: —
| Benchmark | GPT-6 Sol | Phi-3.5-mini |
|---|---|---|
| GDP.pdf | 28% | — |
Reasoning Not comparable
GPT-6 Sol: 49.3 (#34), Phi-3.5-mini: —
| Benchmark | GPT-6 Sol | Phi-3.5-mini |
|---|---|---|
| NYT Connections (extended) | 90.1% | — |
| EBR-Bench | 53.3% | — |
| LMArena Hard Prompts | 1420 | — |
| PIQA | — | 81% |
Math Not comparable
GPT-6 Sol: 35.8 (#135), Phi-3.5-mini: —
| Benchmark | GPT-6 Sol | Phi-3.5-mini |
|---|---|---|
| LMArena Math | 1404 | — |
| GSM8K | — | 86.2% |
Knowledge Not comparable
GPT-6 Sol: 43.0 (#87), Phi-3.5-mini: —
| Benchmark | GPT-6 Sol | Phi-3.5-mini |
|---|---|---|
| Vectara Hallucination Rate | 6.5% | — |
| LMArena Expert | 1449 | — |
| BoolQ | — | 78% |
Multimodal Not comparable
GPT-6 Sol: 47.6 (#9), Phi-3.5-mini: —
| Benchmark | GPT-6 Sol | Phi-3.5-mini |
|---|---|---|
| Furniture Assembly | 58.3% | — |
Multilingual Not comparable
GPT-6 Sol: 51.0 (#113), Phi-3.5-mini: —
| Benchmark | GPT-6 Sol | Phi-3.5-mini |
|---|---|---|
| LMArena Non-English | 1389 | — |
| LMArena Chinese | 1416 | — |
| LMArena Russian | 1407 | — |
| LMArena Spanish | 1377 | — |
Instruction Following Not comparable
GPT-6 Sol: 74.4 (#97), Phi-3.5-mini: —
| Benchmark | GPT-6 Sol | Phi-3.5-mini |
|---|---|---|
| LMArena Instruction Following | 1406 | — |
Long Context Not comparable
GPT-6 Sol: 43.2 (#100), Phi-3.5-mini: —
| Benchmark | GPT-6 Sol | Phi-3.5-mini |
|---|---|---|
| LMArena Longer Query | 1415 | — |
Writing & Preference Not comparable
GPT-6 Sol: 72.3 (#15), Phi-3.5-mini: —
| Benchmark | GPT-6 Sol | Phi-3.5-mini |
|---|---|---|
| LMArena Text | 1396 | — |
| LMArena Creative Writing | 1392 | — |
| EQ-Bench Creative Writing | 2125 | — |
| LMArena Multi-Turn | 1405 | — |
Frequently asked questions
Is GPT-6 Sol better than Phi-3.5-mini?
GPT-6 Sol has enough public results to be ranked (#46); Phi-3.5-mini does not yet, so treat this comparison as directional.
Is GPT-6 Sol or Phi-3.5-mini better for coding?
GPT-6 Sol scores higher on coding benchmarks: 59.6 versus 34.0 in the Noometry coding category.
How many benchmarks do GPT-6 Sol and Phi-3.5-mini share?
0 benchmarks have published results for both models. GPT-6 Sol has 22 scored results on Noometry and Phi-3.5-mini has 5.