Model comparison
DeepSeek-V3.1-Terminus vs Llama 2-7B
DeepSeek-V3.1-Terminus is the stronger model overall, scoring 42.6 to 28.6 on the Noometry Index.
Last verified . 10 shared benchmarks.
DeepSeek
MetaSummary
- They share 10 benchmarks with published results for both. DeepSeek-V3.1-Terminus scores higher in 7 categories and Llama 2-7B in 0 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where DeepSeek-V3.1-Terminus leads 61.3 to 28.1.
Side by side
| DeepSeek-V3.1-Terminus | Llama 2-7B | |
|---|---|---|
| Provider | DeepSeek | Meta |
| Noometry Index | 42.6 | 28.6 |
| Released | 2025-09-22 | 2023-07-18 |
| Weights | Open | Open |
| Context window | 164K | — |
| Max output | 66K | — |
| Input $ / M tokens | $0.30 | — |
| Output $ / M tokens | $1 | — |
| Results tracked | 16 | 29 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 41.5 (#114), Llama 2-7B: 28.7 (#302)
| Benchmark | DeepSeek-V3.1-Terminus | Llama 2-7B |
|---|---|---|
| LMArena Coding | 1426 | 1002 |
| SciCode | 40.6% | — |
| ALE-Bench | 745.17 | — |
Reasoning DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 25.7 (#126), Llama 2-7B: 15.1 (#309)
| Benchmark | DeepSeek-V3.1-Terminus | Llama 2-7B |
|---|---|---|
| LMArena Hard Prompts | 1427 | 1009 |
| Kagi LLM Benchmark | 57.4% | — |
| CritPt | 1.7% | — |
| Chess Puzzles | — | 0% |
| DTBench | 81.3% | — |
| LMCA | 28.6% | — |
| BIG-Bench Hard | — | 39.2% |
| Epoch Capabilities Index | — | 98.54 |
| HellaSwag | — | 77.2% |
| LAMBADA | — | 73.3% |
| PIQA | — | 78.8% |
| WinoGrande | — | 69.2% |
Math DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 35.7 (#138), Llama 2-7B: 28.2 (#231)
| Benchmark | DeepSeek-V3.1-Terminus | Llama 2-7B |
|---|---|---|
| LMArena Math | 1402 | 1042 |
| GSM8K | — | 16.7% |
Knowledge Not comparable
DeepSeek-V3.1-Terminus: —, Llama 2-7B: 27.7 (#243)
| Benchmark | DeepSeek-V3.1-Terminus | Llama 2-7B |
|---|---|---|
| LMArena Expert | — | 1036 |
| ARC (AI2) Challenge | — | 45.9% |
| BoolQ | — | 77.9% |
| MMLU | — | 45.8% |
| OpenBookQA | — | 58.6% |
| TriviaQA | — | 73.7% |
Multimodal Not comparable
DeepSeek-V3.1-Terminus: —, Llama 2-7B: —
| Benchmark | DeepSeek-V3.1-Terminus | Llama 2-7B |
|---|---|---|
| ScienceQA | — | 43.1% |
Multilingual DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 52.3 (#87), Llama 2-7B: 23.9 (#288)
| Benchmark | DeepSeek-V3.1-Terminus | Llama 2-7B |
|---|---|---|
| LMArena Non-English | 1407 | 973 |
| LMArena Russian | 1434 | 995 |
| LMArena Chinese | — | 973 |
| LMArena French | — | 969 |
| LMArena German | — | 978 |
| LMArena Spanish | — | 1007 |
Instruction Following DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 74.4 (#96), Llama 2-7B: 51.0 (#293)
| Benchmark | DeepSeek-V3.1-Terminus | Llama 2-7B |
|---|---|---|
| LMArena Instruction Following | 1406 | 1006 |
Long Context DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 43.5 (#88), Llama 2-7B: 30.4 (#282)
| Benchmark | DeepSeek-V3.1-Terminus | Llama 2-7B |
|---|---|---|
| LMArena Longer Query | 1426 | 999 |
Writing & Preference DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 61.3 (#87), Llama 2-7B: 28.1 (#293)
| Benchmark | DeepSeek-V3.1-Terminus | Llama 2-7B |
|---|---|---|
| LMArena Text | 1420 | 1053 |
| LMArena Creative Writing | 1404 | 1033 |
| LMArena Multi-Turn | 1411 | 1029 |
Frequently asked questions
Is DeepSeek-V3.1-Terminus better than Llama 2-7B?
DeepSeek-V3.1-Terminus is the stronger model overall, scoring 42.6 to 28.6 on the Noometry Index.
Is DeepSeek-V3.1-Terminus or Llama 2-7B better for coding?
DeepSeek-V3.1-Terminus scores higher on coding benchmarks: 41.5 versus 28.7 in the Noometry coding category.
How many benchmarks do DeepSeek-V3.1-Terminus and Llama 2-7B share?
10 benchmarks have published results for both models. DeepSeek-V3.1-Terminus has 16 scored results on Noometry and Llama 2-7B has 29.