Model comparison
Claude 3.5 Haiku vs Grok 4.20 Multi-Agent
Grok 4.20 Multi-Agent is the stronger model overall, scoring 45.5 to 28.9 on the Noometry Index.
Last verified . 18 shared benchmarks.
Summary
- They share 18 benchmarks with published results for both. Claude 3.5 Haiku scores higher in 0 categories and Grok 4.20 Multi-Agent in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.20 Multi-Agent leads 42.4 to 17.1.
Side by side
| Claude 3.5 Haiku | Grok 4.20 Multi-Agent | |
|---|---|---|
| Provider | Anthropic | xAI |
| Noometry Index | 28.9 | 45.5 |
| Released | 2024-10-22 | 2026-03-09 |
| Weights | Proprietary | Proprietary |
| Context window | — | 1M |
| Max output | — | 30K |
| Input $ / M tokens | — | $1.25 |
| Output $ / M tokens | — | $2.50 |
| Results tracked | 49 | 20 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 4.20 Multi-Agent leads
Claude 3.5 Haiku: 32.6 (#258), Grok 4.20 Multi-Agent: 42.4 (#91)
| Benchmark | Claude 3.5 Haiku | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Coding | 1287 | 1458 |
| Aider Polyglot | 28% | — |
| SciCode | 27.4% | — |
| WeirdML | 30.7% | — |
| BigCodeBench Instruct | 46.1% | — |
| LiveBench Coding | 51.4% | — |
| BigCodeBench Complete | 59% | — |
| CadEval | 32% | — |
Agentic & Tool Use Not comparable
Claude 3.5 Haiku: 27.7 (#90), Grok 4.20 Multi-Agent: —
| Benchmark | Claude 3.5 Haiku | Grok 4.20 Multi-Agent |
|---|---|---|
| BALROG | 19.3% | — |
| LMArena Search | — | 1204 |
Reasoning Grok 4.20 Multi-Agent leads
Claude 3.5 Haiku: 17.1 (#286), Grok 4.20 Multi-Agent: 42.4 (#44)
| Benchmark | Claude 3.5 Haiku | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Hard Prompts | 1251 | 1448 |
| NYT Connections (extended) | — | 89.6% |
| CritPt | 0% | — |
| LiveBench Reasoning | 28.1% | — |
| DTBench | 56.7% | — |
| LiveBench Data Analysis | 48.5% | — |
| Epoch Capabilities Index | 127.15 | — |
| LiveBench | 43.5% | — |
Math Grok 4.20 Multi-Agent leads
Claude 3.5 Haiku: 13.7 (#294), Grok 4.20 Multi-Agent: 36.6 (#106)
| Benchmark | Claude 3.5 Haiku | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Math | 1244 | 1441 |
| OTIS Mock AIME 2024-2025 | 4.3% | — |
| Omni-MATH | 22.4% | — |
| LiveBench Math | 35.5% | — |
| MATH Level 5 | 46.4% | — |
| FrontierMath (Feb 2025 set) | 0.3% | — |
Knowledge Grok 4.20 Multi-Agent leads
Claude 3.5 Haiku: 18.7 (#276), Grok 4.20 Multi-Agent: 39.7 (#116)
| Benchmark | Claude 3.5 Haiku | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Expert | 1208 | 1445 |
| GPQA Diamond | 38.1% | — |
| MMLU-Pro | 60.5% | — |
| Confabulations | 36.7% | — |
| GPQA (HELM) | 36.3% | — |
| MMLU | 74.3% | — |
Multimodal Grok 4.20 Multi-Agent leads
Claude 3.5 Haiku: 26.3 (#113), Grok 4.20 Multi-Agent: 39.8 (#44)
| Benchmark | Claude 3.5 Haiku | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Vision | 1092 | 1259 |
| GeoBench | 34% | — |
Multilingual Grok 4.20 Multi-Agent leads
Claude 3.5 Haiku: 40.2 (#213), Grok 4.20 Multi-Agent: 54.6 (#40)
| Benchmark | Claude 3.5 Haiku | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Non-English | 1238 | 1440 |
| LMArena Chinese | 1229 | 1475 |
| LMArena French | 1262 | 1467 |
| LMArena German | 1237 | 1456 |
| LMArena Japanese | 1175 | 1405 |
| LMArena Korean | 1172 | 1416 |
| LMArena Russian | 1253 | 1457 |
| LMArena Spanish | 1261 | 1447 |
Instruction Following Grok 4.20 Multi-Agent leads
Claude 3.5 Haiku: 63.1 (#229), Grok 4.20 Multi-Agent: 75.0 (#80)
| Benchmark | Claude 3.5 Haiku | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Instruction Following | 1241 | 1420 |
| LiveBench Instruction Following | 61.9% | — |
| IFEval | 79.2% | — |
Long Context Grok 4.20 Multi-Agent leads
Claude 3.5 Haiku: 38.3 (#195), Grok 4.20 Multi-Agent: 43.7 (#84)
| Benchmark | Claude 3.5 Haiku | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Longer Query | 1261 | 1430 |
Writing & Preference Grok 4.20 Multi-Agent leads
Claude 3.5 Haiku: 42.9 (#228), Grok 4.20 Multi-Agent: 64.1 (#57)
| Benchmark | Claude 3.5 Haiku | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Text | 1255 | 1450 |
| LMArena Creative Writing | 1233 | 1435 |
| LMArena Multi-Turn | 1265 | 1451 |
| Short-Story Creative Writing | 73.5% | — |
| EQ-Bench Creative Writing | 1146 | — |
| WildBench | 76% | — |
| LiveBench Language | 35.4% | — |
Frequently asked questions
Is Claude 3.5 Haiku better than Grok 4.20 Multi-Agent?
Grok 4.20 Multi-Agent is the stronger model overall, scoring 45.5 to 28.9 on the Noometry Index.
Is Claude 3.5 Haiku or Grok 4.20 Multi-Agent better for coding?
Grok 4.20 Multi-Agent scores higher on coding benchmarks: 42.4 versus 32.6 in the Noometry coding category.
How many benchmarks do Claude 3.5 Haiku and Grok 4.20 Multi-Agent share?
18 benchmarks have published results for both models. Claude 3.5 Haiku has 49 scored results on Noometry and Grok 4.20 Multi-Agent has 20.