Model comparison
Grok 4.20 Multi-Agent vs Phi 3 Mini 4k Instruct
Grok 4.20 Multi-Agent is the stronger model overall, scoring 46.2 to 27.9 on the Noometry Index.
Last verified . 17 shared benchmarks.
MicrosoftSummary
- They share 17 benchmarks with published results for both. Grok 4.20 Multi-Agent scores higher in 8 categories and Phi 3 Mini 4k Instruct in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Grok 4.20 Multi-Agent leads 64.0 to 27.6.
- Phi 3 Mini 4k Instruct has downloadable open weights; the other is API-only.
Side by side
| Grok 4.20 Multi-Agent | Phi 3 Mini 4k Instruct | |
|---|---|---|
| Provider | xAI | Microsoft |
| Noometry Index | 46.2 | 27.9 |
| Released | 2026-03-09 | 2024-04-23 |
| Weights | Proprietary | Open |
| Context window | 1M | — |
| Max output | 30K | — |
| Input $ / M tokens | $1.25 | — |
| Output $ / M tokens | $2.50 | — |
| Results tracked | 20 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 4.20 Multi-Agent leads
Grok 4.20 Multi-Agent: 43.0 (#92), Phi 3 Mini 4k Instruct: 26.6 (#323)
| Benchmark | Grok 4.20 Multi-Agent | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Coding | 1457 | 1093 |
| LiveBench Coding | — | 15.5% |
| HumanEval+ | — | 59.1% |
| MBPP+ | — | 54.2% |
Agentic & Tool Use Not comparable
Grok 4.20 Multi-Agent: —, Phi 3 Mini 4k Instruct: —
| Benchmark | Grok 4.20 Multi-Agent | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Search | 1204 | — |
Reasoning Grok 4.20 Multi-Agent leads
Grok 4.20 Multi-Agent: 43.9 (#48), Phi 3 Mini 4k Instruct: 14.1 (#328)
| Benchmark | Grok 4.20 Multi-Agent | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Hard Prompts | 1448 | 1072 |
| NYT Connections (extended) | 89.6% | — |
| Chess Puzzles | — | 0% |
| LiveBench Reasoning | — | 26.8% |
| LiveBench Data Analysis | — | 34.7% |
| Adversarial NLI | — | 52.8% |
| BIG-Bench Hard | — | 71.7% |
| HellaSwag | — | 76.7% |
| LiveBench | — | 22.4% |
| WinoGrande | — | 70.8% |
Math Grok 4.20 Multi-Agent leads
Grok 4.20 Multi-Agent: 39.4 (#104), Phi 3 Mini 4k Instruct: 26.6 (#257)
| Benchmark | Grok 4.20 Multi-Agent | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Math | 1442 | 1111 |
| LiveBench Math | — | 15.7% |
Knowledge Grok 4.20 Multi-Agent leads
Grok 4.20 Multi-Agent: 40.4 (#119), Phi 3 Mini 4k Instruct: 28.5 (#246)
| Benchmark | Grok 4.20 Multi-Agent | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Expert | 1445 | 1045 |
| ARC (AI2) Challenge | — | 84.9% |
| MMLU | — | 68.8% |
| OpenBookQA | — | 88% |
| TriviaQA | — | 64% |
Multimodal Not comparable
Grok 4.20 Multi-Agent: 40.5 (#48), Phi 3 Mini 4k Instruct: —
| Benchmark | Grok 4.20 Multi-Agent | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Vision | 1259 | — |
Multilingual Grok 4.20 Multi-Agent leads
Grok 4.20 Multi-Agent: 54.4 (#43), Phi 3 Mini 4k Instruct: 26.3 (#280)
| Benchmark | Grok 4.20 Multi-Agent | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Non-English | 1440 | 1021 |
| LMArena Chinese | 1475 | 1021 |
| LMArena French | 1466 | 1076 |
| LMArena German | 1456 | 1044 |
| LMArena Japanese | 1405 | 935 |
| LMArena Korean | 1416 | 905 |
| LMArena Russian | 1457 | 1022 |
| LMArena Spanish | 1447 | 1085 |
Instruction Following Grok 4.20 Multi-Agent leads
Grok 4.20 Multi-Agent: 74.8 (#84), Phi 3 Mini 4k Instruct: 47.7 (#303)
| Benchmark | Grok 4.20 Multi-Agent | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Instruction Following | 1420 | 1053 |
| LiveBench Instruction Following | — | 39.1% |
Long Context Grok 4.20 Multi-Agent leads
Grok 4.20 Multi-Agent: 43.7 (#88), Phi 3 Mini 4k Instruct: 31.7 (#276)
| Benchmark | Grok 4.20 Multi-Agent | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Longer Query | 1431 | 1044 |
Writing & Preference Grok 4.20 Multi-Agent leads
Grok 4.20 Multi-Agent: 64.0 (#59), Phi 3 Mini 4k Instruct: 27.6 (#300)
| Benchmark | Grok 4.20 Multi-Agent | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Text | 1450 | 1073 |
| LMArena Creative Writing | 1436 | 1037 |
| LMArena Multi-Turn | 1452 | 1018 |
| LiveBench Language | — | 9.2% |
Frequently asked questions
Is Grok 4.20 Multi-Agent better than Phi 3 Mini 4k Instruct?
Grok 4.20 Multi-Agent is the stronger model overall, scoring 46.2 to 27.9 on the Noometry Index.
Is Grok 4.20 Multi-Agent or Phi 3 Mini 4k Instruct better for coding?
Grok 4.20 Multi-Agent scores higher on coding benchmarks: 43.0 versus 26.6 in the Noometry coding category.
How many benchmarks do Grok 4.20 Multi-Agent and Phi 3 Mini 4k Instruct share?
17 benchmarks have published results for both models. Grok 4.20 Multi-Agent has 20 scored results on Noometry and Phi 3 Mini 4k Instruct has 35.