LLMs · Zhipu AI
CogVLM2 Llama 3 Chat 19B
- Parameters
- 20B
- Architecture
- Dense
- Max context
- 8K tokens
- License
- cogvlm2
- Released
- 2024-05
How to run it
No mobile app in the Play Store runs CogVLM2 Llama 3 Chat 19B right now: mobile apps need a GGUF version, and none is published. Mobile apps · models that run on mobile devices.
On a mobile device
CogVLM2 Llama 3 Chat 19B is too big for a device like yours: it needs more memory than the device has. Models that run on mobile devices.
Original weights · Hugging Face
Full-precision files from Zhipu AI, for fine-tuning, vLLM, Transformers.
huggingface.co/zai-org/cogvlm2-llama3-chat-19B (opens in a new tab)
Memory needed
Weights plus KV cache and runtime overhead. Leave a little extra free for the operating system and other apps. Try another context length.
| Quantization | Quality | Weights | Total at 8K context | Download |
|---|---|---|---|---|
Q2_K |
Low | 7.8 GB | 12.7 GB | All files (opens in a new tab) |
Q3_K_M |
Low | 9.6 GB | 14.4 GB | All files (opens in a new tab) |
Q4_K_M |
Good | 11.9 GB | 16.8 GB | All files (opens in a new tab) |
Q5_K_M |
High | 13.8 GB | 18.7 GB | All files (opens in a new tab) |
Q6_K |
High | 16.0 GB | 20.9 GB | All files (opens in a new tab) |
Q8_0 |
Very high | 20.7 GB | 25.6 GB | All files (opens in a new tab) |
F16 |
Full | 39.0 GB | 43.9 GB | Original weights (opens in a new tab) |
Best hardware for CogVLM2 Llama 3 Chat 19B
The least expensive GPU in our catalog that runs it without offloading is the AMD Radeon RX 7600 XT (launch price $329) at Q3_K_M.
| Quality | Hardware | Quant | Memory | Speed | Runs |
|---|---|---|---|---|---|
| 38Basic quality (D) · 38 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. |  NVIDIA GeForce RTX 5090 |
Q8_0 | 27.2 GB / 32.0 GB | 57.4 tok/s | Runs great |
| 36Basic quality (D) · 36 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. |  AMD Radeon RX 7900 XTX |
Q5_K_M | 19.9 GB / 24.0 GB | 43.0 tok/s | Runs great |
| 36Basic quality (D) · 36 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. |  NVIDIA GeForce RTX 3090 |
Q5_K_M | 19.9 GB / 24.0 GB | 48.7 tok/s | Runs great |
| 36Basic quality (D) · 36 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. |  NVIDIA GeForce RTX 4090 |
Q5_K_M | 19.9 GB / 24.0 GB | 52.4 tok/s | Runs great |
| 31Basic quality (D) · 31 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. |  AMD Radeon RX 9070 XT |
Q3_K_M | 15.2 GB / 16.0 GB | 44.9 tok/s | Runs welltight fit |
| 31Basic quality (D) · 31 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. |  NVIDIA GeForce RTX 4070 Ti SUPER |
Q3_K_M | 15.2 GB / 16.0 GB | 54.7 tok/s | Runs welltight fit |
| 31Basic quality (D) · 31 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. |  NVIDIA GeForce RTX 4080 SUPER |
Q3_K_M | 15.2 GB / 16.0 GB | 59.9 tok/s | Runs welltight fit |
| 31Basic quality (D) · 31 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. |  NVIDIA GeForce RTX 5070 Ti |
Q3_K_M | 15.2 GB / 16.0 GB | 72.9 tok/s | Runs welltight fit |
| 31Basic quality (D) · 31 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. |  NVIDIA GeForce RTX 5080 |
Q3_K_M | 15.2 GB / 16.0 GB | 78.1 tok/s | Runs welltight fit |
| 31Basic quality (D) · 31 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. |  NVIDIA GeForce RTX 5060 Ti 16GB |
Q3_K_M | 15.2 GB / 16.0 GB | 36.5 tok/s | Runs welltight fit |
Estimates come from published hardware specs and our open formulas, not from running the model on your machine. Real speed varies with runtime, drivers and settings. Numbers marked “Measured” come from a linked benchmark.