Mistral Small 24B Instruct
Mistral · Apache 2.0
Mistral's efficient 24B model with chat, coding, and reasoning capabilities.
- Parameters
- 24B
- Architecture
- Dense
- Context
- 32,768 tokens
- Released
- 2025-01-30
- Engines
- llama.cpp, ollama, vLLM, TGI
- Builder Tools
- Cursor, Continue, Aider, Open WebUI, LM Studio
Parameters
24B
VRAM
20.5 GB
Context
32K
Formats
6
GPUs
20
Mistral Small 24B Instruct (24B) requires 20.5 GB VRAM at recommended quality (Q6_K). At efficient quality (Q4_K_M), it fits in 14 GB VRAM, making it compatible with the AMD Radeon RX 9060 XT 16GB. On NVIDIA Grace Blackwell Ultra GB300, expect approximately 150 tok/s at Q8_0. For the best experience, AMD AI Powerhouse ($1,818) is recommended.
Source: OwnRig methodology
20.5 GB
Q6_K
18 GB
32K tokens
Chat
VRAM Requirements
| Quality | Quantization | VRAM | File Size |
|---|---|---|---|
| full | Q8_0 | 26.5 GB | 24 GB |
| recommended | Q6_K | 20.5 GB | 18 GB |
| recommended | Q5_K_M | 17.2 GB | 15 GB |
| efficient | Q4_K_M | 14 GB | 12 GB |
| compressed | Q3_K_M | 11.2 GB | 9.5 GB |
| compressed | Q2_K | 8.8 GB | 7.4 GB |
Context Length Impact
KV cache VRAM at Q6_K quality. Longer context = more memory.
| Context | KV Cache | Total VRAM |
|---|---|---|
| 2K | 410 MB | 20.9 GB |
| 4K | 819 MB | 21.3 GB |
| 8K | 1.5 GB | 22 GB |
| 16K | 3.1 GB | 23.6 GB |
| 32K | 6.1 GB | 26.6 GBexceeds 24 GB |
Compatible GPUs
20 devices| NVIDIA Grace Blackwell Ultra GB300 | Q8_0 | 150 tok/s | Excellent |
| NVIDIA GeForce RTX 5090 | Q5_K_M | 55 tok/s | Excellent |
| Apple M4 Max (64GB Unified) | Q5_K_M | 22 tok/s | Good |
| NVIDIA GeForce RTX 4090 | Q5_K_M | 32 tok/s | Good |
| AMD Radeon RX 7900 XTX | Q5_K_M | 28 tok/s | Good |
| AMD Radeon Pro W7900 | Q5_K_M | 24 tok/s | Good |
| NVIDIA RTX PRO 6000 Blackwell | Q8_0 | 41 tok/s | Good |
| NVIDIA RTX PRO 6000 Blackwell Max-Q | Q8_0 | 38 tok/s | Good |
| AMD Radeon RX 9070 | Q3_K_M | 32 tok/s | Good |
| NVIDIA GeForce RTX 4070 Ti Super | Q3_K_M | 18 tok/s | Acceptable |
| AMD Radeon RX 9060 XT 16GB | Q3_K_M | 16 tok/s | Acceptable |
| AMD Radeon RX 7600 | Q3_K_M | 2 tok/s | Marginal |
| Apple M3 Pro (18GB Unified) | Q3_K_M | – | Not viable |
| NVIDIA GeForce RTX 3080 10GB | Q2_K | – | Not viable |
| NVIDIA GeForce RTX 4060 8GB | Q3_K_M | – | Not viable |
| NVIDIA RTX 4060 Laptop (40-60W) | Q3_K_M | – | Not viable |
| NVIDIA RTX 4070 Laptop (80-115W) | Q3_K_M | – | Not viable |
| NVIDIA GeForce RTX 4070 Ti 12GB | Q3_K_M | – | Not viable |
| AMD Radeon RX 9060 XT 8GB | Q3_K_M | – | Not viable |
| NVIDIA GeForce RTX 5060 8GB | Q3_K_M | – | Not viable |
Showing 20 of 20 entries
Builder Context
Mistral Small 24B Instruct is commonly used with Cursor, Continue, Aider, Open WebUI, LM Studio. For an AI coding workflow, pair it with an embedding model like nomic-embed-text for local RAG.
Frequently Asked Questions
- How much VRAM does Mistral Small 24B Instruct need?
- Mistral Small 24B Instruct requires 20.5 GB VRAM at recommended quality (Q6_K). At lower quality settings, it can fit in as little as 8.8 GB.
- What is the best GPU for Mistral Small 24B Instruct?
- The NVIDIA Grace Blackwell Ultra GB300 delivers the best performance for Mistral Small 24B Instruct, achieving 150 tok/s at Q8_0 with an excellent rating.
- What quantization should I use for Mistral Small 24B Instruct?
- For the best quality, use Q6_K (20.5 GB VRAM). If your GPU has limited VRAM, Q2_K (8.8 GB) is the most efficient option with acceptable quality.
- Is Mistral Small 24B Instruct good for coding?
- Yes. Mistral Small 24B Instruct is used with Cursor, Continue, Aider, Open WebUI, LM Studio for local AI coding. For the best coding experience, pair it with an embedding model for local RAG.
Data confidence: estimated. Source
VRAM requirements are calculated from model parameters and may vary by inference engine, context length, and batch size. Performance estimates are based on community benchmarks and should be verified for your specific configuration.Mistral is a trademark of its respective owner. OwnRig is not affiliated with or endorsed by the model creator.