Model comparison
DeepSeek V4 Pro vs Llama 2-13B
DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 29.6 on the Noometry Index.
Last verified . 20 shared benchmarks.
DeepSeek
MetaSummary
- They share 20 benchmarks with published results for both. DeepSeek V4 Pro scores higher in 8 categories and Llama 2-13B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where DeepSeek V4 Pro leads 56.5 to 12.8.
- The biggest single-benchmark swing is DTBench: 93.9% for DeepSeek V4 Pro and 42.2% for Llama 2-13B.
Side by side
| DeepSeek V4 Pro | Llama 2-13B | |
|---|---|---|
| Provider | DeepSeek | Meta |
| Noometry Index | 54.3 | 29.6 |
| Released | 2026-04-24 | 2023-07-18 |
| Weights | Open | Open |
| Context window | 1M | — |
| Max output | 393K | — |
| Input $ / M tokens | $0.66 | — |
| Output $ / M tokens | $1.98 | — |
| Results tracked | 48 | 32 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek V4 Pro leads
DeepSeek V4 Pro: 52.4 (#34), Llama 2-13B: 30.9 (#291)
| Benchmark | DeepSeek V4 Pro | Llama 2-13B |
|---|---|---|
| LMArena Coding | 1470 | 1062 |
| SWE-bench Verified | 77.6% | — |
| FrontierCode | 28.6% | — |
| LMArena WebDev | 1582 | — |
| SciCode | 51% | — |
| WeirdML | 66.2% | — |
| ALE-Bench | 1,403 | — |
Agentic & Tool Use Not comparable
DeepSeek V4 Pro: 32.8 (#58), Llama 2-13B: —
| Benchmark | DeepSeek V4 Pro | Llama 2-13B |
|---|---|---|
| APEX-Agents | 47.3% | — |
| Vending-Bench 2 | 3,285 | — |
Reasoning DeepSeek V4 Pro leads
DeepSeek V4 Pro: 56.5 (#24), Llama 2-13B: 12.8 (#337)
| Benchmark | DeepSeek V4 Pro | Llama 2-13B |
|---|---|---|
| Chess Puzzles | 47% | 0% |
| LMArena Hard Prompts | 1461 | 1051 |
| DTBench | 93.9% | 42.2% |
| Epoch Capabilities Index | 155.31 | 106.17 |
| ARC-AGI-2 | 61.3% | — |
| Kagi LLM Benchmark | 53.5% | — |
| NYT Connections (extended) | 91.3% | — |
| ARC-AGI-1 | 90.5% | — |
| CritPt | 18% | — |
| Mystery Game Puzzles | 43% | — |
| LMCA | 45.5% | — |
| Surface Evolver Bench | 40% | — |
| BIG-Bench Hard | — | 58.2% |
| ForecastBench | 56.1 | — |
| HellaSwag | — | 80.7% |
| LAMBADA | — | 76.5% |
| PIQA | — | 80.8% |
| WinoGrande | — | 72.8% |
Math DeepSeek V4 Pro leads
DeepSeek V4 Pro: 64.8 (#30), Llama 2-13B: 31.1 (#229)
| Benchmark | DeepSeek V4 Pro | Llama 2-13B |
|---|---|---|
| LMArena Math | 1455 | 1065 |
| FrontierMath (Tiers 1-3) | 64.6% | — |
| FrontierMath Tier 4 | 26.8% | — |
| MathArena Final-Answer Competitions | 76.6% | — |
| OTIS Mock AIME 2024-2025 | 98.6% | — |
| ProofBench | 50% | — |
| GSM8K | — | 36.9% |
Knowledge DeepSeek V4 Pro leads
DeepSeek V4 Pro: 59.5 (#31), Llama 2-13B: 28.1 (#249)
| Benchmark | DeepSeek V4 Pro | Llama 2-13B |
|---|---|---|
| LMArena Expert | 1464 | 1030 |
| GPQA Diamond | 91.7% | — |
| SimpleQA Verified | 52.9% | — |
| Vectara Hallucination Rate | 8.6% | — |
| ARC (AI2) Challenge | — | 60.3% |
| BoolQ | — | 82.4% |
| MMLU | — | 55.6% |
| OpenBookQA | — | 57% |
| TriviaQA | — | 79.6% |
Multimodal Not comparable
DeepSeek V4 Pro: —, Llama 2-13B: —
| Benchmark | DeepSeek V4 Pro | Llama 2-13B |
|---|---|---|
| ScienceQA | — | 55.8% |
Multilingual DeepSeek V4 Pro leads
DeepSeek V4 Pro: 54.4 (#45), Llama 2-13B: 26.5 (#279)
| Benchmark | DeepSeek V4 Pro | Llama 2-13B |
|---|---|---|
| LMArena Non-English | 1439 | 1024 |
| LMArena Chinese | 1486 | 1001 |
| LMArena French | 1472 | 1044 |
| LMArena German | 1458 | 1009 |
| LMArena Japanese | 1445 | 894 |
| LMArena Korean | 1447 | 953 |
| LMArena Russian | 1453 | 1055 |
| LMArena Spanish | 1458 | 1087 |
Instruction Following DeepSeek V4 Pro leads
DeepSeek V4 Pro: 76.1 (#47), Llama 2-13B: 53.3 (#287)
| Benchmark | DeepSeek V4 Pro | Llama 2-13B |
|---|---|---|
| LMArena Instruction Following | 1448 | 1045 |
Long Context DeepSeek V4 Pro leads
DeepSeek V4 Pro: 45.0 (#51), Llama 2-13B: 32.3 (#269)
| Benchmark | DeepSeek V4 Pro | Llama 2-13B |
|---|---|---|
| LMArena Longer Query | 1458 | 1064 |
| CL-bench Life | 13.5% | — |
Writing & Preference DeepSeek V4 Pro leads
DeepSeek V4 Pro: 65.5 (#46), Llama 2-13B: 29.8 (#289)
| Benchmark | DeepSeek V4 Pro | Llama 2-13B |
|---|---|---|
| LMArena Text | 1451 | 1084 |
| LMArena Creative Writing | 1446 | 1047 |
| LMArena Multi-Turn | 1467 | 1050 |
| EQ-Bench Creative Writing | 1553 | — |
| EQ-Bench 4 | 1166 | — |
Frequently asked questions
Is DeepSeek V4 Pro better than Llama 2-13B?
DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 29.6 on the Noometry Index.
Is DeepSeek V4 Pro or Llama 2-13B better for coding?
DeepSeek V4 Pro scores higher on coding benchmarks: 52.4 versus 30.9 in the Noometry coding category.
How many benchmarks do DeepSeek V4 Pro and Llama 2-13B share?
20 benchmarks have published results for both models. DeepSeek V4 Pro has 48 scored results on Noometry and Llama 2-13B has 32.