- Parameters
- 3.213 B
- Metadata context limit
- —
- Calculation
- Method needed
MAIN CATALOG
Models from leading developers
36 selected models for core tasks. Official releases, without hundreds of conversions and intermediate checkpoints. Popularity does not prove GPU compatibility.
Found: 36
- Parameters
- 8.03 B
- Metadata context limit
- —
- Calculation
- Method needed
- Parameters
- 70.554 B
- Metadata context limit
- —
- Calculation
- Method needed
- Parameters
- 108.642 B
- Metadata context limit
- —
- Calculation
- Method needed
- Parameters
- Unknown
- Metadata context limit
- 262,144
- Calculation
- Method needed
- Parameters
- Unknown
- Metadata context limit
- 1,048,576
- Calculation
- Method needed
- Parameters
- Unknown
- Metadata context limit
- 393,216
- Calculation
- Method needed
- Parameters
- Unknown
- Metadata context limit
- 131,072
- Calculation
- Method needed
- Parameters
- 4.022 B
- Metadata context limit
- 40,960
- Calculation
- Memory lower bound
- Parameters
- 8.191 B
- Metadata context limit
- 40,960
- Calculation
- Memory lower bound
- Parameters
- 32.762 B
- Metadata context limit
- 40,960
- Calculation
- Memory lower bound
- Parameters
- 9.653 B
- Metadata context limit
- 262,144
- Calculation
- Method needed
- Parameters
- 27.781 B
- Metadata context limit
- 262,144
- Calculation
- Method needed
- Parameters
- 8.767 B
- Metadata context limit
- 262,144
- Calculation
- Method needed
- Parameters
- 30.532 B
- Metadata context limit
- 262,144
- Calculation
- Method needed
- Parameters
- 0.596 B
- Metadata context limit
- 32,768
- Calculation
- Method needed
- Parameters
- 1.917 B
- Metadata context limit
- —
- Calculation
- Method needed
- Parameters
- 7.996 B
- Metadata context limit
- 131,072
- Calculation
- Method needed
- Parameters
- 11.96 B
- Metadata context limit
- 262,144
- Calculation
- Method needed
- Parameters
- 31.273 B
- Metadata context limit
- 262,144
- Calculation
- Method needed
- Parameters
- 3.836 B
- Metadata context limit
- 131,072
- Calculation
- Method needed
- Parameters
- 14.66 B
- Metadata context limit
- 16,384
- Calculation
- Method needed
- Parameters
- 8.191 B
- Metadata context limit
- 131,072
- Calculation
- Memory lower bound
- Parameters
- Unknown
- Metadata context limit
- 1,048,576
- Calculation
- Method needed
- Parameters
- Unknown
- Metadata context limit
- 1,048,576
- Calculation
- Method needed
- Parameters
- Unknown
- Metadata context limit
- 131,072
- Calculation
- Method needed
- Parameters
- Unknown
- Metadata context limit
- 131,072
- Calculation
- Method needed
- Parameters
- 0.809 B
- Metadata context limit
- —
- Calculation
- Method needed
- Parameters
- Unknown
- Metadata context limit
- —
- Calculation
- Method needed
- Parameters
- 3.876 B
- Metadata context limit
- —
- Calculation
- Method needed
- Parameters
- 5 B
- Metadata context limit
- —
- Calculation
- Method needed
- Parameters
- 33.123 B
- Metadata context limit
- —
- Calculation
- Method needed
- Parameters
- Unknown
- Metadata context limit
- 204,800
- Calculation
- Method needed
- Parameters
- Unknown
- Metadata context limit
- 1,048,576
- Calculation
- Method needed
- Parameters
- 31.221 B
- Metadata context limit
- 202,752
- Calculation
- Method needed
- Parameters
- Unknown
- Metadata context limit
- 1,048,576
- Calculation
- Method needed
How to plan AI hardware
- How much VRAM does an LLM need?
Estimate LLM memory from weights, quantization, KV cache, context and concurrency. An 8B example and the limits of a preliminary Monsder calculation.
- How to choose an AI server
Choose an AI server from workload requirements: model, load, GPU, memory, power and cooling. What a calculator estimates and what needs engineering validation.
- Which models fit your GPU?
Match an LLM to your GPU using available VRAM, quantization, context and concurrency. Why fitting in memory does not establish a validated deployment.