Key Takeaways
Llama 3.2 11B Vision Instruct wins:
- Cheaper input tokens
- Cheaper output tokens
- Supports vision
Llama 3.3 70B Instruct wins:
- Faster response time
- Higher intelligence benchmark
- Better at coding
- Better at math
Price Advantage
Llama 3.2 11B Vision Instruct
Benchmark Advantage
Llama 3.3 70B Instruct
Context Window
Llama 3.3 70B Instruct
Speed
Llama 3.3 70B Instruct
Pricing Comparison
Price Comparison
| Metric | Llama 3.2 11B Vision Instruct | Llama 3.3 70B Instruct | Winner |
|---|---|---|---|
| Input (per 1M tokens) | $0.05 | $0.10 | Llama 3.2 11B Vision Instruct |
| Output (per 1M tokens) | $0.05 | $0.32 | Llama 3.2 11B Vision Instruct |
Using a 3:1 input/output ratio, Llama 3.2 11B Vision Instruct is 68% cheaper overall.
Llama 3.2 11B Vision Instruct Providers
Cloudflare $0.05 (Cheapest)
DeepInfra $0.05 (Cheapest)
Novita $0.06
Together $0.18
Llama 3.3 70B Instruct Providers
DeepInfra $0.10 (Cheapest)
Novita $0.14
Parasail $0.22
Nebius $0.25
Crusoe $0.25
Benchmark Comparison
8
Benchmarks Compared
0
Llama 3.2 11B Vision Instruct Wins
7
Llama 3.3 70B Instruct Wins
Benchmark Scores
| Benchmark | Llama 3.2 11B Vision Instruct | Llama 3.3 70B Instruct | Winner |
|---|---|---|---|
Intelligence Index Overall intelligence score | 10.9 | 14.2 | |
Coding Index Code generation & understanding | 4.3 | 10.7 | |
Math Index Mathematical reasoning | 1.7 | 7.7 | |
MMLU Pro Academic knowledge | 46.4 | 71.3 | |
GPQA Graduate-level science | 22.1 | 49.8 | |
LiveCodeBench Competitive programming | 11.0 | 28.8 | |
Aider Real-world code editing | - | 59.4 | - |
AIME Competition math | 9.3 | 30.0 |
Llama 3.3 70B Instruct significantly outperforms in coding benchmarks.
Cost vs Quality
X-axis:
Y-axis:
Loading chart...
Llama 3.2 11B Vision Instruct
Other models
Context & Performance
Context Window
Llama 3.2 11B Vision Instruct
131,072
tokens
Max output: 16,384 tokens
Llama 3.3 70B Instruct
131,072
tokens
Max output: 16,384 tokens
Speed Performance
| Metric | Llama 3.2 11B Vision Instruct | Llama 3.3 70B Instruct | Winner |
|---|---|---|---|
| Tokens/second | 69.7 tok/s | 104.4 tok/s | |
| Time to First Token | 0.41s | 0.49s |
Llama 3.3 70B Instruct responds 50% faster on average.
Capabilities
Feature Comparison
| Feature | Llama 3.2 11B Vision Instruct | Llama 3.3 70B Instruct |
|---|---|---|
| Vision (Image Input) | ||
| Tool/Function Calls | ||
| Reasoning Mode | ||
| Audio Input | ||
| Audio Output | ||
| PDF Input | ||
| Prompt Caching | ||
| Web Search |
License & Release
| Property | Llama 3.2 11B Vision Instruct | Llama 3.3 70B Instruct |
|---|---|---|
| License | Open Source | Open Source |
| Author | Meta-llama | Meta-llama |
| Released | Sep 2024 | Dec 2024 |
Llama 3.2 11B Vision Instruct Modalities
Input
textimage
Output
text
Llama 3.3 70B Instruct Modalities
Input
text
Output
text