Model comparison
Grok 4.3 vs phi-3-medium 14B
Grok 4.3 is the stronger model overall, scoring 43.6 to 29.1 on the Noometry Index.
Last verified . 2 shared benchmarks.
MicrosoftSummary
- They share 2 benchmarks with published results for both. Grok 4.3 scores higher in 3 categories and phi-3-medium 14B in 0 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Grok 4.3 leads 52.5 to 9.1.
- The biggest single-benchmark swing is GPQA Diamond: 88.8% for Grok 4.3 and 27.6% for phi-3-medium 14B.
- phi-3-medium 14B has downloadable open weights; the other is API-only.
Side by side
| Grok 4.3 | phi-3-medium 14B | |
|---|---|---|
| Provider | xAI | Microsoft |
| Noometry Index | 43.6 | 29.1 |
| Released | 2026-04-17 | 2024-04-23 |
| Weights | Proprietary | Open |
| Context window | 1M | — |
| Max output | 30K | — |
| Input $ / M tokens | $1.25 | — |
| Output $ / M tokens | $2.50 | — |
| Results tracked | 40 | 13 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 4.3 leads
Grok 4.3: 41.3 (#117), phi-3-medium 14B: 36.3 (#198)
| Benchmark | Grok 4.3 | phi-3-medium 14B |
|---|---|---|
| LMArena WebDev | 1357 | — |
| SciCode | 47.3% | — |
| WeirdML | 49.9% | — |
| BigCodeBench Instruct | — | 37.6% |
| LMArena Coding | 1415 | — |
| BigCodeBench Complete | — | 48.7% |
| ALE-Bench | 944.17 | — |
Agentic & Tool Use Not comparable
Grok 4.3: 27.6 (#93), phi-3-medium 14B: —
| Benchmark | Grok 4.3 | phi-3-medium 14B |
|---|---|---|
| GDP.pdf | 8% | — |
| LMArena Search | 1165 | — |
| Vending-Bench 2 | 35.26 | — |
Reasoning Not comparable
Grok 4.3: 34.9 (#60), phi-3-medium 14B: —
| Benchmark | Grok 4.3 | phi-3-medium 14B |
|---|---|---|
| Epoch Capabilities Index | 149.13 | 121.18 |
| NYT Connections (extended) | 55.2% | — |
| CritPt | 8% | — |
| Chess Puzzles | 25% | — |
| LMArena Hard Prompts | 1397 | — |
| DTBench | 90.7% | — |
| LMCA | 38.3% | — |
| Adversarial NLI | — | 55.8% |
| BIG-Bench Hard | — | 81.4% |
| ForecastBench | 60.3 | — |
| HellaSwag | — | 82.4% |
| WinoGrande | — | 81.5% |
Math Grok 4.3 leads
Grok 4.3: 45.8 (#60), phi-3-medium 14B: 25.2 (#249)
| Benchmark | Grok 4.3 | phi-3-medium 14B |
|---|---|---|
| FrontierMath (Tiers 1-3) | 42.8% | — |
| FrontierMath Tier 4 | 14.6% | — |
| OTIS Mock AIME 2024-2025 | 93.3% | — |
| ProofBench | 11% | — |
| LMArena Math | 1388 | — |
| MATH Level 5 | — | 17.6% |
Knowledge Grok 4.3 leads
Grok 4.3: 52.5 (#57), phi-3-medium 14B: 9.1 (#301)
| Benchmark | Grok 4.3 | phi-3-medium 14B |
|---|---|---|
| GPQA Diamond | 88.8% | 27.6% |
| SimpleQA Verified | 33.2% | — |
| LMArena Expert | 1385 | — |
| ARC (AI2) Challenge | — | 91.6% |
| MMLU | — | 78% |
| OpenBookQA | — | 87.4% |
| TriviaQA | — | 73.9% |
Multimodal Not comparable
Grok 4.3: 31.6 (#96), phi-3-medium 14B: —
| Benchmark | Grok 4.3 | phi-3-medium 14B |
|---|---|---|
| LMArena Vision | 1231 | — |
| Blueprint-Bench 2 | 0% | — |
Multilingual Not comparable
Grok 4.3: 50.7 (#114), phi-3-medium 14B: —
| Benchmark | Grok 4.3 | phi-3-medium 14B |
|---|---|---|
| LMArena Non-English | 1385 | — |
| LMArena Chinese | 1423 | — |
| LMArena French | 1412 | — |
| LMArena German | 1395 | — |
| LMArena Japanese | 1379 | — |
| LMArena Korean | 1357 | — |
| LMArena Russian | 1399 | — |
| LMArena Spanish | 1402 | — |
Instruction Following Not comparable
Grok 4.3: 72.3 (#135), phi-3-medium 14B: —
| Benchmark | Grok 4.3 | phi-3-medium 14B |
|---|---|---|
| LMArena Instruction Following | 1366 | — |
Long Context Not comparable
Grok 4.3: 42.5 (#118), phi-3-medium 14B: —
| Benchmark | Grok 4.3 | phi-3-medium 14B |
|---|---|---|
| LMArena Longer Query | 1393 | — |
Writing & Preference Not comparable
Grok 4.3: 58.6 (#112), phi-3-medium 14B: —
| Benchmark | Grok 4.3 | phi-3-medium 14B |
|---|---|---|
| LMArena Text | 1397 | — |
| LMArena Creative Writing | 1380 | — |
| EQ-Bench 4 | 1075 | — |
| LMArena Multi-Turn | 1406 | — |
Frequently asked questions
Is Grok 4.3 better than phi-3-medium 14B?
Grok 4.3 is the stronger model overall, scoring 43.6 to 29.1 on the Noometry Index.
Is Grok 4.3 or phi-3-medium 14B better for coding?
Grok 4.3 scores higher on coding benchmarks: 41.3 versus 36.3 in the Noometry coding category.
How many benchmarks do Grok 4.3 and phi-3-medium 14B share?
2 benchmarks have published results for both models. Grok 4.3 has 40 scored results on Noometry and phi-3-medium 14B has 13.