Model comparison
Claude 3 Haiku vs Granite 4.1 8b
Granite 4.1 8b is the stronger model overall, scoring 37.4 to 25.9 on the Noometry Index.
Last verified . 12 shared benchmarks.
Summary
- They share 12 benchmarks with published results for both. Claude 3 Haiku scores higher in 0 categories and Granite 4.1 8b in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Granite 4.1 8b leads 36.4 to 9.8.
- Granite 4.1 8b has downloadable open weights; the other is API-only.
Side by side
| Claude 3 Haiku | Granite 4.1 8b | |
|---|---|---|
| Provider | Anthropic | IBM |
| Noometry Index | 25.9 | 37.4 |
| Released | 2024-03-07 | — |
| Weights | Proprietary | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 37 | 13 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Granite 4.1 8b leads
Claude 3 Haiku: 26.4 (#325), Granite 4.1 8b: 30.1 (#297)
| Benchmark | Claude 3 Haiku | Granite 4.1 8b |
|---|---|---|
| LMArena Coding | 1199 | 1312 |
| LMArena WebDev | — | 1192 |
| WeirdML | 9.8% | — |
| BigCodeBench Instruct | 39.4% | — |
| BigCodeBench Complete | 50.1% | — |
| CadEval | 12% | — |
| HumanEval+ | 68.9% | — |
| MBPP+ | 68.8% | — |
Reasoning Granite 4.1 8b leads
Claude 3 Haiku: 16.3 (#307), Granite 4.1 8b: 25.7 (#143)
| Benchmark | Claude 3 Haiku | Granite 4.1 8b |
|---|---|---|
| LMArena Hard Prompts | 1174 | 1293 |
| Kagi LLM Benchmark | 34.2% | — |
| DTBench | 50.1% | — |
| LMCA | 8.8% | — |
| Epoch Capabilities Index | 118.35 | — |
| ForecastBench | 53.2 | — |
| WinoGrande | 74.2% | — |
Math Granite 4.1 8b leads
Claude 3 Haiku: 9.8 (#319), Granite 4.1 8b: 36.4 (#166)
| Benchmark | Claude 3 Haiku | Granite 4.1 8b |
|---|---|---|
| LMArena Math | 1188 | 1312 |
| OTIS Mock AIME 2024-2025 | 1.8% | — |
| MATH Level 5 | 14.9% | — |
Knowledge Granite 4.1 8b leads
Claude 3 Haiku: 17.3 (#285), Granite 4.1 8b: 36.1 (#174)
| Benchmark | Claude 3 Haiku | Granite 4.1 8b |
|---|---|---|
| LMArena Expert | 1148 | 1309 |
| GPQA Diamond | 36.3% | — |
| Confabulations | 34.2% | — |
| MMLU | 73.8% | — |
Multimodal Not comparable
Claude 3 Haiku: 23.6 (#128), Granite 4.1 8b: —
| Benchmark | Claude 3 Haiku | Granite 4.1 8b |
|---|---|---|
| LMArena Vision | 950 | — |
| ScienceQA | 72% | — |
Multilingual Granite 4.1 8b leads
Claude 3 Haiku: 36.0 (#243), Granite 4.1 8b: 41.7 (#204)
| Benchmark | Claude 3 Haiku | Granite 4.1 8b |
|---|---|---|
| LMArena Non-English | 1178 | 1261 |
| LMArena Chinese | 1155 | 1337 |
| LMArena Russian | 1204 | 1240 |
| LMArena French | 1195 | — |
| LMArena German | 1174 | — |
| LMArena Japanese | 1102 | — |
| LMArena Korean | 1109 | — |
| LMArena Spanish | 1166 | — |
Instruction Following Granite 4.1 8b leads
Claude 3 Haiku: 61.3 (#247), Granite 4.1 8b: 66.9 (#203)
| Benchmark | Claude 3 Haiku | Granite 4.1 8b |
|---|---|---|
| LMArena Instruction Following | 1173 | 1269 |
Long Context Granite 4.1 8b leads
Claude 3 Haiku: 36.1 (#237), Granite 4.1 8b: 38.7 (#193)
| Benchmark | Claude 3 Haiku | Granite 4.1 8b |
|---|---|---|
| LMArena Longer Query | 1190 | 1275 |
Writing & Preference Granite 4.1 8b leads
Claude 3 Haiku: 29.7 (#291), Granite 4.1 8b: 48.1 (#204)
| Benchmark | Claude 3 Haiku | Granite 4.1 8b |
|---|---|---|
| LMArena Text | 1195 | 1290 |
| LMArena Creative Writing | 1157 | 1251 |
| LMArena Multi-Turn | 1190 | 1268 |
| EQ-Bench Creative Writing | 717 | — |
Frequently asked questions
Is Claude 3 Haiku better than Granite 4.1 8b?
Granite 4.1 8b is the stronger model overall, scoring 37.4 to 25.9 on the Noometry Index.
Is Claude 3 Haiku or Granite 4.1 8b better for coding?
Granite 4.1 8b scores higher on coding benchmarks: 30.1 versus 26.4 in the Noometry coding category.
How many benchmarks do Claude 3 Haiku and Granite 4.1 8b share?
12 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and Granite 4.1 8b has 13.