Model comparison
DeepSeek-V2.5 (Sep 2024) vs Llama 3.1-8B
DeepSeek-V2.5 (Sep 2024) is the stronger model overall, scoring 37.6 to 23.0 on the Noometry Index.
Last verified . 21 shared benchmarks.
DeepSeek
MetaSummary
- They share 21 benchmarks with published results for both. DeepSeek-V2.5 (Sep 2024) scores higher in 8 categories and Llama 3.1-8B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where DeepSeek-V2.5 (Sep 2024) leads 34.8 to 8.0.
- The biggest single-benchmark swing is BigCodeBench Instruct: 48.6% for DeepSeek-V2.5 (Sep 2024) and 32.8% for Llama 3.1-8B.
Side by side
| DeepSeek-V2.5 (Sep 2024) | Llama 3.1-8B | |
|---|---|---|
| Provider | DeepSeek | Meta |
| Noometry Index | 37.6 | 23.0 |
| Released | 2024-09-06 | 2024-07-23 |
| Weights | Open | Open |
| Context window | — | 128K |
| Max output | — | 4K |
| Input $ / M tokens | — | $0.05 |
| Output $ / M tokens | — | $0.08 |
| Results tracked | 22 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 31.7 (#281), Llama 3.1-8B: 20.2 (#340)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Llama 3.1-8B |
|---|---|---|
| BigCodeBench Instruct | 48.6% | 32.8% |
| LMArena Coding | 1309 | 1195 |
| BigCodeBench Complete | 53.2% | 40.5% |
| HumanEval+ | 83.5% | 62.8% |
| MBPP+ | 74.1% | 55.6% |
| Aider Polyglot | 17.8% | — |
| SciCode | — | 13.2% |
| WeirdML | — | 1.7% |
Agentic & Tool Use Not comparable
DeepSeek-V2.5 (Sep 2024): —, Llama 3.1-8B: 22.5 (#131)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Llama 3.1-8B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 25.8% |
| BALROG | — | 15.1% |
Reasoning DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 25.6 (#145), Llama 3.1-8B: 14.9 (#321)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Llama 3.1-8B |
|---|---|---|
| LMArena Hard Prompts | 1289 | 1175 |
| CritPt | — | 0% |
| Chess Puzzles | — | 0% |
| DTBench | — | 50.9% |
| LMCA | — | 5.4% |
| Epoch Capabilities Index | — | 116.57 |
| PIQA | — | 81.2% |
Math DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 35.9 (#177), Llama 3.1-8B: 10.2 (#317)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Llama 3.1-8B |
|---|---|---|
| LMArena Math | 1288 | 1179 |
| OTIS Mock AIME 2024-2025 | — | 1.7% |
| Omni-MATH | — | 13.7% |
| MATH Level 5 | — | 22.9% |
| GSM8K | — | 82.4% |
Knowledge DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 34.8 (#193), Llama 3.1-8B: 8.0 (#307)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Llama 3.1-8B |
|---|---|---|
| LMArena Expert | 1266 | 1144 |
| GPQA Diamond | — | 27% |
| MMLU-Pro | — | 40.6% |
| GPQA (HELM) | — | 24.7% |
| BoolQ | — | 82.8% |
| MMLU | — | 56.1% |
Multilingual DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 42.5 (#193), Llama 3.1-8B: 34.0 (#249)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Llama 3.1-8B |
|---|---|---|
| LMArena Non-English | 1273 | 1148 |
| LMArena Chinese | 1318 | 1151 |
| LMArena French | 1289 | 1177 |
| LMArena German | 1258 | 1144 |
| LMArena Japanese | 1228 | 1061 |
| LMArena Korean | 1209 | 1053 |
| LMArena Russian | 1289 | 1158 |
| LMArena Spanish | 1248 | 1169 |
Instruction Following DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 67.5 (#194), Llama 3.1-8B: 58.9 (#258)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Llama 3.1-8B |
|---|---|---|
| LMArena Instruction Following | 1280 | 1159 |
| IFEval | — | 74.3% |
Long Context DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 39.5 (#174), Llama 3.1-8B: 35.8 (#238)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Llama 3.1-8B |
|---|---|---|
| LMArena Longer Query | 1301 | 1182 |
Writing & Preference DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 49.8 (#187), Llama 3.1-8B: 29.7 (#290)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Llama 3.1-8B |
|---|---|---|
| LMArena Text | 1294 | 1187 |
| LMArena Creative Writing | 1285 | 1154 |
| LMArena Multi-Turn | 1297 | 1172 |
| EQ-Bench Creative Writing | — | 713 |
| WildBench | — | 68.7% |
Frequently asked questions
Is DeepSeek-V2.5 (Sep 2024) better than Llama 3.1-8B?
DeepSeek-V2.5 (Sep 2024) is the stronger model overall, scoring 37.6 to 23.0 on the Noometry Index.
Is DeepSeek-V2.5 (Sep 2024) or Llama 3.1-8B better for coding?
DeepSeek-V2.5 (Sep 2024) scores higher on coding benchmarks: 31.7 versus 20.2 in the Noometry coding category.
How many benchmarks do DeepSeek-V2.5 (Sep 2024) and Llama 3.1-8B share?
21 benchmarks have published results for both models. DeepSeek-V2.5 (Sep 2024) has 22 scored results on Noometry and Llama 3.1-8B has 43.