DeepSeek V3 (Dec '24)
DeepSeek V3 (Dec '24) is a text-based AI model by DeepSeek. It features 0 tok/s output speed, $0.40/1M input tokens pricing. It scores 16.5 on the intelligence index.
Strengths
- +Very affordable pricing ($0.40/1M tokens)
Weaknesses
- -Slower than average output (0 tok/s)
Benchmarks
Intelligence16.5Coding16.4Math26.0
MMLU-Pro
0.8
GPQA
0.6
LiveCodeBench
0.4
Intelligence
16.5
Coding
16.4
Math
26.0
Output Speed
0 tok/s
Latency (TTFT)
0.00s
Blended Price
$0.63/M
Pricing per 1M Tokens
| Input (Prompt) | $0.40 |
| Output (Completion) | $0.89 |
| Cache Read | Free |
| Cache Write | Free |
Specifications
| Context Length | N/A |
| Max Output Tokens | N/A |
| Input Modalities | Text |
| Output Modalities | Text |
| Tokenizer | N/A |
Compare DeepSeek V3 (Dec '24) with:
Alternatives to DeepSeek V3 (Dec '24)
More from DeepSeek
Frequently Asked Questions
How much does DeepSeek V3 (Dec '24) cost?
DeepSeek V3 (Dec '24) costs $0.40/1M input tokens and $0.89/1M output tokens.
Is DeepSeek V3 (Dec '24) good for coding?
DeepSeek V3 (Dec '24) has a lower coding score of 16.4. For demanding coding tasks, consider a model with a higher coding benchmark.
How fast is DeepSeek V3 (Dec '24)?
DeepSeek V3 (Dec '24) generates output at 0 tok/s. Time to first token is 0.00s.
Is there a free version of DeepSeek V3 (Dec '24)?
No, DeepSeek V3 (Dec '24) is a paid model. Check the free models page for zero-cost alternatives.
What are alternatives to DeepSeek V3 (Dec '24)?
Last updated: