Model comparison
Gemini 3.8 Flash vs Llama 2-70B
Gemini 3.8 Flash is the stronger model overall, scoring 61.8 to 24.4 on the Noometry Index.
Last verified . 21 shared benchmarks.
Google
MetaSummary
- They share 21 benchmarks with published results for both. Gemini 3.8 Flash scores higher in 8 categories and Llama 2-70B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Gemini 3.8 Flash leads 74.8 to 7.4.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 98.9% for Gemini 3.8 Flash and 0% for Llama 2-70B.
- Llama 2-70B has downloadable open weights; the other is API-only.
Side by side
| Gemini 3.8 Flash | Llama 2-70B | |
|---|---|---|
| Provider | Meta | |
| Noometry Index | 61.8 | 24.4 |
| Released | 2026-09-02 | 2023-07-18 |
| Weights | Proprietary | Open |
| Context window | 1.05M | — |
| Max output | 66K | — |
| Input $ / M tokens | $0.75 | — |
| Output $ / M tokens | $3.75 | — |
| Results tracked | 50 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemini 3.8 Flash leads
Gemini 3.8 Flash: 59.2 (#15), Llama 2-70B: 31.4 (#286)
| Benchmark | Gemini 3.8 Flash | Llama 2-70B |
|---|---|---|
| LMArena Coding | 1510 | 1079 |
| DeepSWE | 73.8% | — |
| FrontierCode | 41.2% | — |
| CursorBench | 39.6% | — |
| LMArena WebDev | 1584 | — |
| FrontierSWE | 19.6% | — |
| SciCode | 56.6% | — |
| WeirdML | 84.8% | — |
| ALE-Bench | 1,270 | — |
Agentic & Tool Use Not comparable
Gemini 3.8 Flash: 41.8 (#21), Llama 2-70B: —
| Benchmark | Gemini 3.8 Flash | Llama 2-70B |
|---|---|---|
| APEX-Agents | 64.3% | — |
| Remote Labor Index | 5.8% | — |
| GDP.pdf | 23.4% | — |
| Vending-Bench 2 | 5,094 | — |
Reasoning Gemini 3.8 Flash leads
Gemini 3.8 Flash: 76.9 (#5), Llama 2-70B: 14.4 (#325)
| Benchmark | Gemini 3.8 Flash | Llama 2-70B |
|---|---|---|
| LMArena Hard Prompts | 1508 | 1073 |
| DTBench | 95.7% | 41.6% |
| Epoch Capabilities Index | 156.71 | 113.79 |
| ARC-AGI-2 | 89.2% | — |
| NYT Connections (extended) | 97.4% | — |
| ARC-AGI-1 | 98.5% | — |
| CritPt | 18.3% | — |
| Chess Puzzles | 61% | — |
| Mystery Game Puzzles | 47% | — |
| LMCA | 52.9% | — |
| Surface Evolver Bench | 76.9% | — |
| BIG-Bench Hard | — | 64.9% |
| CommonsenseQA 2.0 | — | 50% |
| ForecastBench | — | 51.4 |
| HellaSwag | — | 85.3% |
| LAMBADA | — | 78.9% |
| PIQA | — | 82.8% |
| WinoGrande | — | 80.2% |
Math Gemini 3.8 Flash leads
Gemini 3.8 Flash: 65.3 (#28), Llama 2-70B: 8.1 (#326)
| Benchmark | Gemini 3.8 Flash | Llama 2-70B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 98.9% | 0% |
| LMArena Math | 1528 | 1091 |
| FrontierMath (Tiers 1-3) | 68.4% | — |
| FrontierMath Tier 4 | 22% | — |
| ProofBench | 48% | — |
| MATH Level 5 | — | 3.3% |
| GSM8K | — | 69.6% |
Knowledge Gemini 3.8 Flash leads
Gemini 3.8 Flash: 74.8 (#2), Llama 2-70B: 7.4 (#310)
| Benchmark | Gemini 3.8 Flash | Llama 2-70B |
|---|---|---|
| GPQA Diamond | 95.4% | 26.3% |
| LMArena Expert | 1524 | 1039 |
| Humanity's Last Exam | 44.5% | — |
| SimpleQA Verified | 69.7% | — |
| ARC (AI2) Challenge | — | 78.3% |
| BoolQ | — | 88.6% |
| MMLU | — | 69.9% |
| OpenBookQA | — | 60.2% |
| TriviaQA | — | 87.6% |
Multimodal Not comparable
Gemini 3.8 Flash: 40.7 (#45), Llama 2-70B: —
| Benchmark | Gemini 3.8 Flash | Llama 2-70B |
|---|---|---|
| LMArena Vision | 1314 | — |
| Blueprint-Bench 2 | 38.6% | — |
| Furniture Assembly | 31.7% | — |
Multilingual Gemini 3.8 Flash leads
Gemini 3.8 Flash: 58.0 (#5), Llama 2-70B: 27.7 (#274)
| Benchmark | Gemini 3.8 Flash | Llama 2-70B |
|---|---|---|
| LMArena Non-English | 1491 | 1045 |
| LMArena Chinese | 1554 | 995 |
| LMArena French | 1498 | 1090 |
| LMArena German | 1493 | 1041 |
| LMArena Japanese | 1502 | 927 |
| LMArena Korean | 1459 | 964 |
| LMArena Russian | 1515 | 1083 |
| LMArena Spanish | 1485 | 1143 |
Instruction Following Gemini 3.8 Flash leads
Gemini 3.8 Flash: 78.0 (#13), Llama 2-70B: 54.9 (#278)
| Benchmark | Gemini 3.8 Flash | Llama 2-70B |
|---|---|---|
| LMArena Instruction Following | 1490 | 1071 |
Long Context Gemini 3.8 Flash leads
Gemini 3.8 Flash: 46.3 (#24), Llama 2-70B: 32.3 (#270)
| Benchmark | Gemini 3.8 Flash | Llama 2-70B |
|---|---|---|
| LMArena Longer Query | 1508 | 1062 |
Writing & Preference Gemini 3.8 Flash leads
Gemini 3.8 Flash: 72.2 (#15), Llama 2-70B: 32.3 (#279)
| Benchmark | Gemini 3.8 Flash | Llama 2-70B |
|---|---|---|
| LMArena Text | 1499 | 1115 |
| LMArena Creative Writing | 1492 | 1075 |
| LMArena Multi-Turn | 1501 | 1088 |
| EQ-Bench Creative Writing | 1748 | — |
Frequently asked questions
Is Gemini 3.8 Flash better than Llama 2-70B?
Gemini 3.8 Flash is the stronger model overall, scoring 61.8 to 24.4 on the Noometry Index.
Is Gemini 3.8 Flash or Llama 2-70B better for coding?
Gemini 3.8 Flash scores higher on coding benchmarks: 59.2 versus 31.4 in the Noometry coding category.
How many benchmarks do Gemini 3.8 Flash and Llama 2-70B share?
21 benchmarks have published results for both models. Gemini 3.8 Flash has 50 scored results on Noometry and Llama 2-70B has 35.