Ollama Library · llama
Can I run reflection?
A high-performing model trained with a new technique called Reflection-tuning that teaches a LLM to detect mistakes in its reasoning and correct course.
70.6B parameters128K contexttextLLAMA 3.1 COMMUNITY LICENSE AGREEMENT6 builds
reflection system requirements
Memory needed at 4K context, counting model weights, KV cache, compute buffers and runtime overhead. Download sizes come from the official Ollama registry.
| Build | Quantization | Download | Memory needed |
|---|---|---|---|
| reflection:70b | Q4_0 | 37.2 GB | ~53.6 GB(est.) |
| reflection:70b-q4_K_M | Q4_K_M | 39.6 GB | ~57.0 GB(est.) |
| reflection:70b-q5_K_M | Q5_K_M | 46.5 GB | ~66.9 GB(est.) |
| reflection:70b-q6_K | Q6_K | 53.9 GB | ~77.5 GB(est.) |
| reflection:70b-q8_0 | Q8_0 | 69.8 GB | ~100.2 GB(est.) |
| reflection:70b-fp16 | F16 | 131.4 GB | ~188.3 GB(est.) |
Figures marked (est.) approximate the KV cache because this model's architecture details are not published. They are indicative rather than exact.
How to run reflection locally
- 1.Install Ollama for Windows, macOS or Linux.
- 2.Run this in a terminal:
ollama run reflection:70b - 3.The weights download on first run, then an interactive prompt opens.
Frequently asked questions about reflection
- How much VRAM do I need to run reflection?
- At 4K context, the smallest published build of reflection (Q4_0) needs roughly 53.6 GB of GPU memory once weights, KV cache, compute buffers and runtime overhead are counted. Bigger quantizations and longer context windows need more. It can also run partly or entirely on the CPU using system RAM, more slowly.
- Can I run reflection without a dedicated GPU?
- Yes, but slowly. Without a GPU the model runs on the CPU using system RAM, which usually means a few tokens per second rather than dozens. You would need at least 53.6 GB of free RAM for the smallest build.
- How big is the reflection download?
- The smallest published build is 37.2 GB. There are 6 builds in total, the largest being 131.4 GB. Leave some extra free disk space beyond the download itself.
- How do I run reflection locally?
- Install Ollama, then run "ollama run reflection:70b" in a terminal. The weights download on first use and an interactive session opens.
Will reflection run on your PC?
Scan your hardware once and get a verdict for this model and every build — free, no account.
Model data from the official Ollama registry · verified 13/09/2026