An advanced language model crafted with 2 trillion bilingual tokens.
ollama run deepseek-llm
Ollama kurulu değil mi? ollama.com/download — Windows, macOS ve Linux için ücretsiz. İlk çalıştırmada model indirilir, sonrası tamamen çevrimdışıdır.
Boyut büyüdükçe kalite artar, donanım ihtiyacı yükselir. Başlangıç için küçük varyantı deneyin.
| Etiket | Boyut | Bağlam | Girdi | Komut |
|---|---|---|---|---|
| latest | 4.0GB | 4K | Text |
ollama run deepseek-llm:latest
|
| 7b | 4.0GB | 4K | Text |
ollama run deepseek-llm:7b
|
| 67b | 38GB | 4K | Text |
ollama run deepseek-llm:67b
|
| 7b-base | 4.0GB | 4K | Text |
ollama run deepseek-llm:7b-base
|
| 7b-base-q2_K | 3.0GB | 4K | Text |
ollama run deepseek-llm:7b-base-q2_K
|
| 7b-base-q3_K_S | 3.1GB | 4K | Text |
ollama run deepseek-llm:7b-base-q3_K_S
|
| 7b-base-q3_K_M | 3.5GB | 4K | Text |
ollama run deepseek-llm:7b-base-q3_K_M
|
| 7b-base-q3_K_L | 3.7GB | 4K | Text |
ollama run deepseek-llm:7b-base-q3_K_L
|
| 7b-base-q4_0 | 4.0GB | 4K | Text |
ollama run deepseek-llm:7b-base-q4_0
|
| 7b-base-q4_1 | 4.4GB | 4K | Text |
ollama run deepseek-llm:7b-base-q4_1
|
| 7b-base-q4_K_S | 4.0GB | 4K | Text |
ollama run deepseek-llm:7b-base-q4_K_S
|
| 7b-base-q4_K_M | 4.2GB | 4K | Text |
ollama run deepseek-llm:7b-base-q4_K_M
|
| 7b-base-q5_0 | 4.8GB | 4K | Text |
ollama run deepseek-llm:7b-base-q5_0
|
| 7b-base-q5_1 | 5.2GB | 4K | Text |
ollama run deepseek-llm:7b-base-q5_1
|
| 7b-base-q5_K_S | 4.8GB | 4K | Text |
ollama run deepseek-llm:7b-base-q5_K_S
|
| 7b-base-q5_K_M | 4.9GB | 4K | Text |
ollama run deepseek-llm:7b-base-q5_K_M
|
| 7b-base-q6_K | 5.7GB | 4K | Text |
ollama run deepseek-llm:7b-base-q6_K
|
| 7b-base-q8_0 | 7.3GB | 4K | Text |
ollama run deepseek-llm:7b-base-q8_0
|
| 7b-base-fp16 | 14GB | 4K | Text |
ollama run deepseek-llm:7b-base-fp16
|
| 7b-chat | 4.0GB | 4K | Text |
ollama run deepseek-llm:7b-chat
|
| 7b-chat-q2_K | 3.0GB | 4K | Text |
ollama run deepseek-llm:7b-chat-q2_K
|
| 7b-chat-q3_K_S | 3.1GB | 4K | Text |
ollama run deepseek-llm:7b-chat-q3_K_S
|
| 7b-chat-q3_K_M | 3.5GB | 4K | Text |
ollama run deepseek-llm:7b-chat-q3_K_M
|
| 7b-chat-q3_K_L | 3.7GB | 4K | Text |
ollama run deepseek-llm:7b-chat-q3_K_L
|
| 7b-chat-q4_0 | 4.0GB | 4K | Text |
ollama run deepseek-llm:7b-chat-q4_0
|
39 varyant daha var — tümü için ollama.com.
DeepSeek LLM is an advanced language model available in both 7 billion and 67 billion parameters. Both a chat and base variation are available.
Superior General Capabilities: DeepSeek LLM 67B Base outperforms Llama2 70B Base in areas such as reasoning, coding, math, and Chinese comprehension.
Proficient in Coding and Math: DeepSeek LLM 67B Chat exhibits outstanding performance in coding (using the HumanEval benchmark) and mathematics (using the GSM8K benchmark).