Key Takeaways
Qwen3 30B A3B Instruct 2507 wins:
- Larger context window
- Faster response time
- Better at coding
- Better at math
- Supports tool calls
Llama 3 8B Lunaris wins:
- Cheaper input tokens
- Cheaper output tokens
- Higher intelligence benchmark
- Has reasoning mode
Price Advantage
Llama 3 8B Lunaris
Benchmark Advantage
Qwen3 30B A3B Instruct 2507
Context Window
Qwen3 30B A3B Instruct 2507
Speed
Qwen3 30B A3B Instruct 2507
Pricing Comparison
Price Comparison
| Metric | Qwen3 30B A3B Instruct 2507 | Llama 3 8B Lunaris | Winner |
|---|---|---|---|
| Input (per 1M tokens) | $0.09 | $0.04 | Llama 3 8B Lunaris |
| Output (per 1M tokens) | $0.30 | $0.05 | Llama 3 8B Lunaris |
| Cache Read (per 1M) | $0.08 | N/A | Qwen3 30B A3B Instruct 2507 |
Using a 3:1 input/output ratio, Llama 3 8B Lunaris is 70% cheaper overall.
Qwen3 30B A3B Instruct 2507 Providers
No provider data available
Llama 3 8B Lunaris Providers
No provider data available
Benchmark Comparison
8
Benchmarks Compared
2
Qwen3 30B A3B Instruct 2507 Wins
1
Llama 3 8B Lunaris Wins
Benchmark Scores
| Benchmark | Qwen3 30B A3B Instruct 2507 | Llama 3 8B Lunaris | Winner |
|---|---|---|---|
Intelligence Index Overall intelligence score | 15.0 | 25.6 | |
Coding Index Code generation & understanding | 14.2 | - | - |
Math Index Mathematical reasoning | 66.3 | - | - |
MMLU Pro Academic knowledge | 77.7 | 31.0 | |
GPQA Graduate-level science | 65.9 | 6.8 | |
LiveCodeBench Competitive programming | 51.5 | - | - |
AIME Competition math | 72.7 | - | - |
BBH Big-Bench Hard | - | 32.1 | - |
Qwen3 30B A3B Instruct 2507 significantly outperforms in coding benchmarks.
Cost vs Quality
X-axis:
Y-axis:
Loading chart...
Qwen3 30B A3B Instruct 2507
Other models
Context & Performance
Context Window
Qwen3 30B A3B Instruct 2507
262,144
tokens
Llama 3 8B Lunaris
8,192
tokens
Qwen3 30B A3B Instruct 2507 has a 97% larger context window.
Speed Performance
| Metric | Qwen3 30B A3B Instruct 2507 | Llama 3 8B Lunaris | Winner |
|---|---|---|---|
| Tokens/second | 82.2 tok/s | N/A | |
| Time to First Token | 0.98s | N/A |
Capabilities
Feature Comparison
| Feature | Qwen3 30B A3B Instruct 2507 | Llama 3 8B Lunaris |
|---|---|---|
| Vision (Image Input) | ||
| Tool/Function Calls | ||
| Reasoning Mode | ||
| Audio Input | ||
| Audio Output | ||
| PDF Input | ||
| Prompt Caching | ||
| Web Search |
License & Release
| Property | Qwen3 30B A3B Instruct 2507 | Llama 3 8B Lunaris |
|---|---|---|
| License | Open Source | Open Source |
| Author | Qwen | Sao10k |
| Released | Jul 2025 | Aug 2024 |
Qwen3 30B A3B Instruct 2507 Modalities
Input
text
Output
text
Llama 3 8B Lunaris Modalities
Input
text
Output
text