Model comparison
Claude Haiku 4.5 vs Grok-2 (Dec 2024)
Claude Haiku 4.5 is the stronger model overall, scoring 39.5 to 33.7 on the Noometry Index.
Last verified . 24 shared benchmarks.
Summary
- They share 24 benchmarks with published results for both. Claude Haiku 4.5 scores higher in 7 categories and Grok-2 (Dec 2024) in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Haiku 4.5 leads 44.9 to 20.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 66.7% for Claude Haiku 4.5 and 11.5% for Grok-2 (Dec 2024).
Side by side
| Claude Haiku 4.5 | Grok-2 (Dec 2024) | |
|---|---|---|
| Provider | Anthropic | xAI |
| Noometry Index | 39.5 | 33.7 |
| Released | 2025-10-15 | 2024-08-13 |
| Weights | Proprietary | Proprietary |
| Context window | 200K | — |
| Max output | 64K | — |
| Input $ / M tokens | $1 | — |
| Output $ / M tokens | $5 | — |
| Results tracked | 53 | 34 |
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: 44.0 (#78), Grok-2 (Dec 2024): 33.3 (#258)
| Benchmark | Claude Haiku 4.5 | Grok-2 (Dec 2024) |
|---|---|---|
| WeirdML | 45.4% | 22.2% |
| LMArena Coding | 1453 | 1287 |
| SWE-bench Verified (bash only) | 66.6% | — |
| LMArena WebDev | 1330 | — |
| SWE-bench Multilingual | 64.7% | — |
| SciCode | 43.3% | — |
| LiveBench Coding | — | 46.4% |
| ALE-Bench | 653.48 | — |
Agentic & Tool Use Not comparable
Claude Haiku 4.5: 33.6 (#52), Grok-2 (Dec 2024): —
| Benchmark | Claude Haiku 4.5 | Grok-2 (Dec 2024) |
|---|---|---|
| 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 Grok-2 (Dec 2024) leads
Claude Haiku 4.5: 15.1 (#320), Grok-2 (Dec 2024): 16.9 (#299)
| Benchmark | Claude Haiku 4.5 | Grok-2 (Dec 2024) |
|---|---|---|
| LMArena Hard Prompts | 1420 | 1272 |
| DTBench | 73.6% | 65.2% |
| Epoch Capabilities Index | 142.41 | 130.48 |
| ARC-AGI-2 | 4% | — |
| SimpleBench | — | 22.7% |
| NYT Connections (extended) | 14.3% | — |
| ARC-AGI-1 | 47.7% | — |
| CritPt | 0% | — |
| Chess Puzzles | 8% | — |
| LiveBench Reasoning | — | 54.8% |
| LiveBench Data Analysis | — | 54.5% |
| LMCA | 30.9% | — |
| ForecastBench | 61.4 | — |
| LiveBench | — | 54.3% |
Math Claude Haiku 4.5 leads
Claude Haiku 4.5: 44.9 (#78), Grok-2 (Dec 2024): 20.8 (#284)
| Benchmark | Claude Haiku 4.5 | Grok-2 (Dec 2024) |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 66.7% | 11.5% |
| LMArena Math | 1396 | 1283 |
| MATH Level 5 | 96.4% | 63.5% |
| FrontierMath (Feb 2025 set) | 5.9% | 0.7% |
| Omni-MATH | 56.1% | — |
| LiveBench Math | — | 54.9% |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge Claude Haiku 4.5 leads
Claude Haiku 4.5: 37.7 (#153), Grok-2 (Dec 2024): 29.8 (#233)
| Benchmark | Claude Haiku 4.5 | Grok-2 (Dec 2024) |
|---|---|---|
| GPQA Diamond | 71.2% | 53.8% |
| LMArena Expert | 1442 | 1254 |
| SimpleQA Verified | 13.2% | — |
| MMLU-Pro | 77.7% | — |
| Confabulations | — | 20.1% |
| Vectara Hallucination Rate | 9.8% | — |
| GPQA (HELM) | 60.5% | — |
Multimodal Not comparable
Claude Haiku 4.5: 26.8 (#118), Grok-2 (Dec 2024): —
| Benchmark | Claude Haiku 4.5 | Grok-2 (Dec 2024) |
|---|---|---|
| Blueprint-Bench 2 | 0% | — |
| LMArena Document | 1420 | — |
Multilingual Claude Haiku 4.5 leads
Claude Haiku 4.5: 49.9 (#129), Grok-2 (Dec 2024): 43.1 (#188)
| Benchmark | Claude Haiku 4.5 | Grok-2 (Dec 2024) |
|---|---|---|
| LMArena Non-English | 1377 | 1282 |
| LMArena Chinese | 1417 | 1289 |
| LMArena French | 1408 | 1318 |
| LMArena German | 1375 | 1287 |
| LMArena Japanese | 1339 | 1244 |
| LMArena Korean | 1347 | 1237 |
| LMArena Russian | 1381 | 1286 |
| LMArena Spanish | 1420 | 1281 |
Instruction Following Claude Haiku 4.5 leads
Claude Haiku 4.5: 71.4 (#149), Grok-2 (Dec 2024): 66.9 (#202)
| Benchmark | Claude Haiku 4.5 | Grok-2 (Dec 2024) |
|---|---|---|
| LMArena Instruction Following | 1414 | 1270 |
| LiveBench Instruction Following | — | 69.6% |
| IFEval | 80.1% | — |
Long Context Claude Haiku 4.5 leads
Claude Haiku 4.5: 43.6 (#92), Grok-2 (Dec 2024): 38.8 (#190)
| Benchmark | Claude Haiku 4.5 | Grok-2 (Dec 2024) |
|---|---|---|
| LMArena Longer Query | 1427 | 1276 |
Writing & Preference Claude Haiku 4.5 leads
Claude Haiku 4.5: 57.9 (#123), Grok-2 (Dec 2024): 48.6 (#198)
| Benchmark | Claude Haiku 4.5 | Grok-2 (Dec 2024) |
|---|---|---|
| LMArena Text | 1396 | 1305 |
| LMArena Creative Writing | 1372 | 1284 |
| LMArena Multi-Turn | 1409 | 1290 |
| Short-Story Creative Writing | — | 63.6% |
| WildBench | 83.9% | — |
| EQ-Bench 4 | 1064 | — |
| LiveBench Language | — | 45.6% |
Frequently asked questions
Is Claude Haiku 4.5 better than Grok-2 (Dec 2024)?
Claude Haiku 4.5 is the stronger model overall, scoring 39.5 to 33.7 on the Noometry Index.
Is Claude Haiku 4.5 or Grok-2 (Dec 2024) better for coding?
Claude Haiku 4.5 scores higher on coding benchmarks: 44.0 versus 33.3 in the Noometry coding category.
How many benchmarks do Claude Haiku 4.5 and Grok-2 (Dec 2024) share?
24 benchmarks have published results for both models. Claude Haiku 4.5 has 53 scored results on Noometry and Grok-2 (Dec 2024) has 34.