Model comparison
Granite 3.1 8b Instruct vs Llama 3.1-70B
Granite 3.1 8b Instruct is the stronger model overall, scoring 32.4 to 29.6 on the Noometry Index.
Last verified . 12 shared benchmarks.
MetaSummary
- They share 12 benchmarks with published results for both. Granite 3.1 8b Instruct scores higher in 5 categories and Llama 3.1-70B in 4 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in math, where Granite 3.1 8b Instruct leads 33.0 to 13.5.
Side by side
| Granite 3.1 8b Instruct | Llama 3.1-70B | |
|---|---|---|
| Provider | IBM | Meta |
| Noometry Index | 32.4 | 29.6 |
| Released | — | 2024-07-23 |
| Weights | Open | Open |
| Context window | — | 128K |
| Max output | — | 4K |
| Input $ / M tokens | — | $0.40 |
| Output $ / M tokens | — | $0.40 |
| Results tracked | 13 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Granite 3.1 8b Instruct leads
Granite 3.1 8b Instruct: 34.5 (#233), Llama 3.1-70B: 30.3 (#296)
| Benchmark | Granite 3.1 8b Instruct | Llama 3.1-70B |
|---|---|---|
| LMArena Coding | 1186 | 1260 |
| WeirdML | — | 9% |
| BigCodeBench Instruct | — | 46.1% |
| BigCodeBench Complete | — | 54.8% |
Agentic & Tool Use Too close to call
Granite 3.1 8b Instruct: 24.1 (#120), Llama 3.1-70B: 25.1 (#112)
| Benchmark | Granite 3.1 8b Instruct | Llama 3.1-70B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 27.1% | — |
| TheAgentCompany | — | 6.9% |
| BALROG | — | 27.9% |
Reasoning Too close to call
Granite 3.1 8b Instruct: 22.1 (#207), Llama 3.1-70B: 21.6 (#220)
| Benchmark | Granite 3.1 8b Instruct | Llama 3.1-70B |
|---|---|---|
| LMArena Hard Prompts | 1145 | 1241 |
| DTBench | — | 60% |
| LMCA | — | 14.8% |
| Epoch Capabilities Index | — | 125.92 |
Math Granite 3.1 8b Instruct leads
Granite 3.1 8b Instruct: 33.0 (#209), Llama 3.1-70B: 13.5 (#304)
| Benchmark | Granite 3.1 8b Instruct | Llama 3.1-70B |
|---|---|---|
| LMArena Math | 1152 | 1252 |
| OTIS Mock AIME 2024-2025 | — | 3.6% |
| Omni-MATH | — | 21% |
| MATH Level 5 | — | 36.7% |
Knowledge Granite 3.1 8b Instruct leads
Granite 3.1 8b Instruct: 31.1 (#220), Llama 3.1-70B: 24.2 (#269)
| Benchmark | Granite 3.1 8b Instruct | Llama 3.1-70B |
|---|---|---|
| LMArena Expert | 1142 | 1209 |
| GPQA Diamond | — | 44.2% |
| MMLU-Pro | — | 65.3% |
| GPQA (HELM) | — | 42.6% |
| MMLU | — | 80.1% |
Multilingual Llama 3.1-70B leads
Granite 3.1 8b Instruct: 30.9 (#260), Llama 3.1-70B: 38.8 (#225)
| Benchmark | Granite 3.1 8b Instruct | Llama 3.1-70B |
|---|---|---|
| LMArena Non-English | 1099 | 1219 |
| LMArena Chinese | 1145 | 1215 |
| LMArena Russian | 1092 | 1234 |
| LMArena French | — | 1261 |
| LMArena German | — | 1222 |
| LMArena Japanese | — | 1132 |
| LMArena Korean | — | 1140 |
| LMArena Spanish | — | 1253 |
Instruction Following Llama 3.1-70B leads
Granite 3.1 8b Instruct: 58.6 (#259), Llama 3.1-70B: 65.3 (#223)
| Benchmark | Granite 3.1 8b Instruct | Llama 3.1-70B |
|---|---|---|
| LMArena Instruction Following | 1131 | 1231 |
| IFEval | — | 82.1% |
Long Context Llama 3.1-70B leads
Granite 3.1 8b Instruct: 35.2 (#241), Llama 3.1-70B: 37.6 (#214)
| Benchmark | Granite 3.1 8b Instruct | Llama 3.1-70B |
|---|---|---|
| LMArena Longer Query | 1162 | 1241 |
Writing & Preference Too close to call
Granite 3.1 8b Instruct: 35.5 (#266), Llama 3.1-70B: 35.4 (#267)
| Benchmark | Granite 3.1 8b Instruct | Llama 3.1-70B |
|---|---|---|
| LMArena Text | 1150 | 1261 |
| LMArena Creative Writing | 1129 | 1232 |
| LMArena Multi-Turn | 1108 | 1256 |
| EQ-Bench Creative Writing | — | 784 |
| WildBench | — | 75.8% |
Frequently asked questions
Is Granite 3.1 8b Instruct better than Llama 3.1-70B?
Granite 3.1 8b Instruct is the stronger model overall, scoring 32.4 to 29.6 on the Noometry Index.
Is Granite 3.1 8b Instruct or Llama 3.1-70B better for coding?
Granite 3.1 8b Instruct scores higher on coding benchmarks: 34.5 versus 30.3 in the Noometry coding category.
How many benchmarks do Granite 3.1 8b Instruct and Llama 3.1-70B share?
12 benchmarks have published results for both models. Granite 3.1 8b Instruct has 13 scored results on Noometry and Llama 3.1-70B has 35.