Model comparison
Gemini 3 Deep Think vs Llama 3.2 3B
Llama 3.2 3B has enough public results to be ranked (#318); Gemini 3 Deep Think does not yet, so treat this comparison as directional.
Last verified . 0 shared benchmarks.
Google
MetaSummary
- The widest gap is in reasoning, where Gemini 3 Deep Think leads 70.8 to 20.1.
- Llama 3.2 3B has downloadable open weights; the other is API-only.
Side by side
| Gemini 3 Deep Think | Llama 3.2 3B | |
|---|---|---|
| Provider | Meta | |
| Noometry Index | 58.5 | 28.4 |
| Released | 2026-02-12 | 2024-09-24 |
| Weights | Proprietary | Open |
| Context window | — | 131K |
| Max output | — | 118K |
| Input $ / M tokens | — | $0.05 |
| Output $ / M tokens | — | $0.33 |
| Results tracked | 3 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Gemini 3 Deep Think: —, Llama 3.2 3B: 27.1 (#314)
| Benchmark | Gemini 3 Deep Think | Llama 3.2 3B |
|---|---|---|
| BigCodeBench Instruct | — | 23.4% |
| LMArena Coding | — | 1098 |
| BigCodeBench Complete | — | 28.3% |
Agentic & Tool Use Not comparable
Gemini 3 Deep Think: —, Llama 3.2 3B: 19.9 (#135)
| Benchmark | Gemini 3 Deep Think | Llama 3.2 3B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 21.9% |
| BALROG | — | 10.1% |
Reasoning Gemini 3 Deep Think leads
Gemini 3 Deep Think: 70.8 (#11), Llama 3.2 3B: 20.1 (#225)
| Benchmark | Gemini 3 Deep Think | Llama 3.2 3B |
|---|---|---|
| ARC-AGI-2 | 84.6% | — |
| ARC-AGI-1 | 96% | — |
| CritPt | 25.7% | — |
| LMArena Hard Prompts | — | 1095 |
Math Not comparable
Gemini 3 Deep Think: —, Llama 3.2 3B: 29.9 (#213)
| Benchmark | Gemini 3 Deep Think | Llama 3.2 3B |
|---|---|---|
| LMArena Math | — | 1126 |
Knowledge Not comparable
Gemini 3 Deep Think: —, Llama 3.2 3B: 29.1 (#232)
| Benchmark | Gemini 3 Deep Think | Llama 3.2 3B |
|---|---|---|
| LMArena Expert | — | 1090 |
Multilingual Not comparable
Gemini 3 Deep Think: —, Llama 3.2 3B: 26.4 (#276)
| Benchmark | Gemini 3 Deep Think | Llama 3.2 3B |
|---|---|---|
| LMArena Non-English | — | 1019 |
| LMArena Chinese | — | 1017 |
| LMArena German | — | 1056 |
| LMArena Russian | — | 949 |
Instruction Following Not comparable
Gemini 3 Deep Think: —, Llama 3.2 3B: 56.3 (#270)
| Benchmark | Gemini 3 Deep Think | Llama 3.2 3B |
|---|---|---|
| LMArena Instruction Following | — | 1089 |
Long Context Not comparable
Gemini 3 Deep Think: —, Llama 3.2 3B: 33.4 (#256)
| Benchmark | Gemini 3 Deep Think | Llama 3.2 3B |
|---|---|---|
| LMArena Longer Query | — | 1101 |
Writing & Preference Not comparable
Gemini 3 Deep Think: —, Llama 3.2 3B: 24.9 (#302)
| Benchmark | Gemini 3 Deep Think | Llama 3.2 3B |
|---|---|---|
| LMArena Text | — | 1110 |
| LMArena Creative Writing | — | 1094 |
| EQ-Bench Creative Writing | — | 595 |
| LMArena Multi-Turn | — | 1105 |
Frequently asked questions
Is Gemini 3 Deep Think better than Llama 3.2 3B?
Llama 3.2 3B has enough public results to be ranked (#318); Gemini 3 Deep Think does not yet, so treat this comparison as directional.
How many benchmarks do Gemini 3 Deep Think and Llama 3.2 3B share?
0 benchmarks have published results for both models. Gemini 3 Deep Think has 3 scored results on Noometry and Llama 3.2 3B has 18.