Sidebar
tips:llm
LLM
For Production Deployment:
- Primary Choice: DeepSeek-R1 32B for reasoning-heavy applications
- Coding Tasks: Qwen2.5-Coder 7B for optimal balance of capability and efficiency
- General Purpose: Llama 3.3 70B for maximum versatility
- Edge Computing: Phi-4 14B for resource-constrained environments
Optimization Strategies:
- Always enable Flash Attention and KV-cache quantization
- Use Q4_K_M quantization for production deployments
- Implement caching for repeated queries
- Monitor GPU memory usage and implement automatic model swapping
- Use load balancing for high-throughput applications
| Hardware | Llama 3.3 8B (tokens/sec) | Llama 3.3 70B (tokens/sec) | Llama 3.2 |
|---|---|---|---|
| RTX 4090 | 89.2 | 12.1 | |
| RTX 3090 | 67.4 | 8.3 | |
| A100 40GB | 156.7 | 45.2 | |
| M3 Max 128GB | 34.8 | 4.2 | |
| Strix Halo 128GB ollama | 5.1 | 85.02 | |
| Strix Halo 128GB llama.cpp | 90 | ||
| RTX 3060 | 131.76 |
ROCM
| model | capabilities | size | context | quantization | eval rate [token/s] | prompt eval rate [token/s] |
|---|---|---|---|---|---|---|
| llama3.2 | completion tools | “3.2B” | 131072 | “Q4KM” | 52.78 | 1957.30 |
| qwen-strixhalo | completion tools | “30.5B” | 262144 | “Q4KM” | 53.54 | 1056.37 |
| qwen3-coder | completion tools | “30.5B” | 262144 | “Q4KM” | 52.10 | 776.55 |
| qwen3:30b-a3b | completion tools thinking | “30.5B” | 262144 | “Q4KM” | 50.19 | 803.06 |
| gpt-oss:20b | completion tools thinking | “20.9B” | 131072 | “MXFP4” | 45.37 | 519.90 |
| glm-4.7-flash | completion tools thinking | “29.9B” | 202752 | “Q4KM” | 41.54 | 470.09 |
| qwen3:8b | completion tools thinking | “8.2B” | 40960 | “Q4KM” | 32.68 | 890.98 |
| qwen3-coder-next | completion tools | “79.7B” | 262144 | “Q4KM” | 33.06 | 380.21 |
| qwen2.5-coder:14b-instruct-q4KM | completion tools insert | “14.8B” | 32768 | “Q4_K_M” | 17.25 | 527.74 |
VULKAN
| model | capabilities | size | context | quantization | eval rate [token/s] | prompt eval rate [token/s] |
|---|---|---|---|---|---|---|
| qwen3-coder | completion tools | “30.5B” | 262144 | “Q4KM” | 54.03 | 805.43 |
| llama3.2 | completion tools | “3.2B” | 131072 | “Q4_K_M” | 52.54 | 1838.82 |
| gpt-oss:20b | completion tools thinking | “20.9B” | 131072 | “MXFP4” | 43.36 | 475.60 |
ollama model
FROM qwen3-coder # STRIX HALO AGENTIC TUNING PARAMETER num_ctx 128000 PARAMETER num_batch 1024 PARAMETER num_predict 4096 SYSTEM """ You are a Strix Halo Optimized Coding Agent. Always use asynchronous patterns and favor memory-efficient algorithms. """
tips/llm.txt · Last modified: by sscipioni
