Model comparison
DeepSeek V4.1 Flash vs Llama 2-13B
DeepSeek V4.1 Flash is the stronger model overall, scoring 52.8 to 29.6 on the Noometry Index.
Last verified . 19 shared benchmarks.
DeepSeek
MetaSummary
- They share 19 benchmarks with published results for both. DeepSeek V4.1 Flash 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 V4.1 Flash leads 50.2 to 12.8.
- The biggest single-benchmark swing is DTBench: 89.9% for DeepSeek V4.1 Flash and 42.2% for Llama 2-13B.
Side by side
| DeepSeek V4.1 Flash | Llama 2-13B | |
|---|---|---|
| Provider | DeepSeek | Meta |
| Noometry Index | 52.8 | 29.6 |
| Released | 2026-09-09 | 2023-07-18 |
| Weights | Open | Open |
| Context window | 1M | — |
| Max output | 393K | — |
| Input $ / M tokens | $0.15 | — |
| Output $ / M tokens | $0.60 | — |
| Results tracked | 37 | 32 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 52.9 (#32), Llama 2-13B: 30.9 (#291)
| Benchmark | DeepSeek V4.1 Flash | Llama 2-13B |
|---|---|---|
| LMArena Coding | 1506 | 1062 |
| LMArena WebDev | 1619 | — |
| SciCode | 51.9% | — |
| ALE-Bench | 1,092 | — |
Agentic & Tool Use Not comparable
DeepSeek V4.1 Flash: 31.2 (#69), Llama 2-13B: —
| Benchmark | DeepSeek V4.1 Flash | Llama 2-13B |
|---|---|---|
| APEX-Agents | 39.5% | — |
| GDP.pdf | 19.8% | — |
Reasoning DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 50.2 (#36), Llama 2-13B: 12.8 (#337)
| Benchmark | DeepSeek V4.1 Flash | Llama 2-13B |
|---|---|---|
| LMArena Hard Prompts | 1483 | 1051 |
| DTBench | 89.9% | 42.2% |
| Epoch Capabilities Index | 154.9 | 106.17 |
| NYT Connections (extended) | 89.6% | — |
| CritPt | 14.3% | — |
| Chess Puzzles | — | 0% |
| Mystery Game Puzzles | 43% | — |
| LMCA | 47% | — |
| Surface Evolver Bench | 46.3% | — |
| BIG-Bench Hard | — | 58.2% |
| HellaSwag | — | 80.7% |
| LAMBADA | — | 76.5% |
| PIQA | — | 80.8% |
| WinoGrande | — | 72.8% |
Math DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 66.7 (#25), Llama 2-13B: 31.1 (#229)
| Benchmark | DeepSeek V4.1 Flash | Llama 2-13B |
|---|---|---|
| LMArena Math | 1477 | 1065 |
| FrontierMath (Tiers 1-3) | 67.4% | — |
| FrontierMath Tier 4 | 26.8% | — |
| OTIS Mock AIME 2024-2025 | 98.3% | — |
| ProofBench | 54% | — |
| GSM8K | — | 36.9% |
Knowledge DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 57.9 (#38), Llama 2-13B: 28.1 (#249)
| Benchmark | DeepSeek V4.1 Flash | Llama 2-13B |
|---|---|---|
| LMArena Expert | 1506 | 1030 |
| GPQA Diamond | 89.8% | — |
| ARC (AI2) Challenge | — | 60.3% |
| BoolQ | — | 82.4% |
| MMLU | — | 55.6% |
| OpenBookQA | — | 57% |
| TriviaQA | — | 79.6% |
Multimodal Not comparable
DeepSeek V4.1 Flash: 39.1 (#61), Llama 2-13B: —
| Benchmark | DeepSeek V4.1 Flash | Llama 2-13B |
|---|---|---|
| LMArena Vision | 1277 | — |
| Furniture Assembly | 34.2% | — |
| ScienceQA | — | 55.8% |
Multilingual DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 55.0 (#35), Llama 2-13B: 26.5 (#279)
| Benchmark | DeepSeek V4.1 Flash | Llama 2-13B |
|---|---|---|
| LMArena Non-English | 1448 | 1024 |
| LMArena Chinese | 1497 | 1001 |
| LMArena French | 1452 | 1044 |
| LMArena German | 1484 | 1009 |
| LMArena Japanese | 1412 | 894 |
| LMArena Korean | 1452 | 953 |
| LMArena Russian | 1471 | 1055 |
| LMArena Spanish | 1459 | 1087 |
Instruction Following DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 77.3 (#26), Llama 2-13B: 53.3 (#287)
| Benchmark | DeepSeek V4.1 Flash | Llama 2-13B |
|---|---|---|
| LMArena Instruction Following | 1474 | 1045 |
Long Context DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 45.2 (#47), Llama 2-13B: 32.3 (#269)
| Benchmark | DeepSeek V4.1 Flash | Llama 2-13B |
|---|---|---|
| LMArena Longer Query | 1475 | 1064 |
Writing & Preference DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 65.4 (#48), Llama 2-13B: 29.8 (#289)
| Benchmark | DeepSeek V4.1 Flash | Llama 2-13B |
|---|---|---|
| LMArena Text | 1462 | 1084 |
| LMArena Creative Writing | 1435 | 1047 |
| LMArena Multi-Turn | 1457 | 1050 |
| EQ-Bench Creative Writing | 1540 | — |
Frequently asked questions
Is DeepSeek V4.1 Flash better than Llama 2-13B?
DeepSeek V4.1 Flash is the stronger model overall, scoring 52.8 to 29.6 on the Noometry Index.
Is DeepSeek V4.1 Flash or Llama 2-13B better for coding?
DeepSeek V4.1 Flash scores higher on coding benchmarks: 52.9 versus 30.9 in the Noometry coding category.
How many benchmarks do DeepSeek V4.1 Flash and Llama 2-13B share?
19 benchmarks have published results for both models. DeepSeek V4.1 Flash has 37 scored results on Noometry and Llama 2-13B has 32.