DeepSeek V3 0324
DeepSeek chat model for instruction following, coding, and analysis
- Strongest at
- Writing Better than 42% of the past year's models · 2 of 4 benchmarks
- Price per 1M tokens
- $0.29 / $1.14 input / output · OpenRouter
- Context window
- 163.8K tokens in · 147.5K out
- Since the last one
- 88 days after DeepSeek-V3
Capability profile
Where it lands among 90 models released in the past year. 50 is the average model.
Where these scores come from
Each capability combines the published benchmarks listed here, rescaled so 50 is the reference average and 15 points is one standard deviation. A hollow marker rests on a single benchmark.
- Writing EQ-Bench Creative Writing v3, EQ-Bench Longform Writing
- Science WeirdML v2
Price against capability
The shaded corner holds anything better and cheaper. Prices blend input and output 3 to 1.
4 models from the past year score higher on Writing for less: DeepSeek V4 Flash Vision Exp, DeepSeek V4.1 Flash, DeepSeek V4 Pro and 1 more.
15 models from the past year score higher on Science for less: DeepSeek V4.1 Flash 1017, GPT-6 Luna, GLM-5.3-Flash and 12 more.
OpenRouter $0.29 / $1.14 · Cached input $0.11
Since DeepSeek-V3
8 releases in the DeepSeek V4 Preview line.
A newer release is out: DeepSeek V4 Preview.
DeepSeek's past year
9 release days in the past year, about one every 41 days. The next is expected around Nov 25, 2026. Forecasts ↗
Specifications
Specifications and prices from Models.dev; availability and limits may differ by provider. Benchmarks from Epoch AI (CC BY 4.0), LiveBench and the other sources named above.