Model comparison
Claude Haiku 4.5 vs Grok-3 mini
Grok-3 mini is the stronger model overall, scoring 41.0 to 39.3 on the Noometry Index.
Last verified . 30 shared benchmarks.
Summary
- They share 30 benchmarks with published results for both. Claude Haiku 4.5 scores higher in 6 categories and Grok-3 mini in 2 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Grok-3 mini leads 46.3 to 37.6.
- The biggest single-benchmark swing is ARC-AGI-1: 47.7% for Claude Haiku 4.5 and 16.5% for Grok-3 mini.
Side by side
| Claude Haiku 4.5 | Grok-3 mini | |
|---|---|---|
| Provider | Anthropic | xAI |
| Noometry Index | 39.3 | 41.0 |
| Released | 2025-10-15 | 2025-04-09 |
| Weights | Proprietary | Proprietary |
| Context window | 200K | — |
| Max output | 64K | — |
| Input $ / M tokens | $1 | — |
| Output $ / M tokens | $5 | — |
| Results tracked | 53 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Haiku 4.5 leads
Claude Haiku 4.5: 43.9 (#71), Grok-3 mini: 40.4 (#126)
| Benchmark | Claude Haiku 4.5 | Grok-3 mini |
|---|---|---|
| WeirdML | 45.4% | 42.6% |
| LMArena Coding | 1454 | 1380 |
| SWE-bench Verified (bash only) | 66.6% | — |
| Aider Polyglot | — | 49.3% |
| LMArena WebDev | 1330 | — |
| SWE-bench Multilingual | 64.7% | — |
| SciCode | 43.3% | — |
| ALE-Bench | 653.48 | — |
Agentic & Tool Use Not comparable
Claude Haiku 4.5: 33.5 (#50), Grok-3 mini: —
| Benchmark | Claude Haiku 4.5 | Grok-3 mini |
|---|---|---|
| Terminal-Bench | 35.5% | — |
| Berkeley Function Calling Leaderboard | 68.7% | — |
| DeepResearch Bench | 45.5% | — |
| BALROG | 31.2% | — |
| ExploitBench | 13.7% | — |
| Vending-Bench 2 | 458.89 | — |
Reasoning Claude Haiku 4.5 leads
Claude Haiku 4.5: 15.1 (#310), Grok-3 mini: 13.8 (#323)
| Benchmark | Claude Haiku 4.5 | Grok-3 mini |
|---|---|---|
| ARC-AGI-2 | 4% | 0.4% |
| ARC-AGI-1 | 47.7% | 16.5% |
| LMArena Hard Prompts | 1420 | 1374 |
| Epoch Capabilities Index | 142.42 | 140.35 |
| Kagi LLM Benchmark | — | 61.3% |
| NYT Connections (extended) | 14.3% | — |
| CritPt | 0% | — |
| Chess Puzzles | 8% | — |
| DTBench | 73.6% | — |
| LMCA | 30.9% | — |
| ForecastBench | 61.4 | — |
Math Claude Haiku 4.5 leads
Claude Haiku 4.5: 43.0 (#69), Grok-3 mini: 40.3 (#78)
| Benchmark | Claude Haiku 4.5 | Grok-3 mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 66.7% | 77.8% |
| Omni-MATH | 56.1% | 31.8% |
| LMArena Math | 1395 | 1386 |
| MATH Level 5 | 96.4% | 90.9% |
| FrontierMath (Feb 2025 set) | 5.9% | 5.9% |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge Grok-3 mini leads
Claude Haiku 4.5: 37.6 (#146), Grok-3 mini: 46.3 (#75)
| Benchmark | Claude Haiku 4.5 | Grok-3 mini |
|---|---|---|
| GPQA Diamond | 71.2% | 76.3% |
| MMLU-Pro | 77.7% | 79.9% |
| GPQA (HELM) | 60.5% | 67.5% |
| LMArena Expert | 1443 | 1395 |
| SimpleQA Verified | 13.2% | — |
| Confabulations | — | 10.8% |
| Vectara Hallucination Rate | 9.8% | — |
Multimodal Not comparable
Claude Haiku 4.5: 27.0 (#111), Grok-3 mini: —
| Benchmark | Claude Haiku 4.5 | Grok-3 mini |
|---|---|---|
| Blueprint-Bench 2 | 0% | — |
| LMArena Document | 1420 | — |
Multilingual Claude Haiku 4.5 leads
Claude Haiku 4.5: 50.1 (#124), Grok-3 mini: 48.3 (#141)
| Benchmark | Claude Haiku 4.5 | Grok-3 mini |
|---|---|---|
| LMArena Non-English | 1376 | 1352 |
| LMArena Chinese | 1415 | 1387 |
| LMArena French | 1408 | 1357 |
| LMArena German | 1377 | 1350 |
| LMArena Japanese | 1339 | 1343 |
| LMArena Korean | 1346 | 1336 |
| LMArena Russian | 1380 | 1352 |
| LMArena Spanish | 1420 | 1379 |
Instruction Following Grok-3 mini leads
Claude Haiku 4.5: 71.6 (#144), Grok-3 mini: 78.7 (#8)
| Benchmark | Claude Haiku 4.5 | Grok-3 mini |
|---|---|---|
| IFEval | 80.1% | 95.1% |
| LMArena Instruction Following | 1414 | 1356 |
Long Context Claude Haiku 4.5 leads
Claude Haiku 4.5: 43.6 (#86), Grok-3 mini: 41.0 (#142)
| Benchmark | Claude Haiku 4.5 | Grok-3 mini |
|---|---|---|
| LMArena Longer Query | 1427 | 1372 |
| Fiction.LiveBench | — | 66.7% |
Writing & Preference Claude Haiku 4.5 leads
Claude Haiku 4.5: 58.1 (#119), Grok-3 mini: 52.7 (#163)
| Benchmark | Claude Haiku 4.5 | Grok-3 mini |
|---|---|---|
| LMArena Text | 1396 | 1370 |
| LMArena Creative Writing | 1371 | 1342 |
| WildBench | 83.9% | 65.1% |
| LMArena Multi-Turn | 1409 | 1355 |
| Short-Story Creative Writing | — | 73.5% |
| EQ-Bench 4 | 1064 | — |
Frequently asked questions
Is Claude Haiku 4.5 better than Grok-3 mini?
Grok-3 mini is the stronger model overall, scoring 41.0 to 39.3 on the Noometry Index.
Is Claude Haiku 4.5 or Grok-3 mini better for coding?
Claude Haiku 4.5 scores higher on coding benchmarks: 43.9 versus 40.4 in the Noometry coding category.
How many benchmarks do Claude Haiku 4.5 and Grok-3 mini share?
30 benchmarks have published results for both models. Claude Haiku 4.5 has 53 scored results on Noometry and Grok-3 mini has 35.