Sulat.com
AI models
$10 off the fastest DeepSeek V4.1 Flash, Kimi K3 and GLM 5.3 from Synthetic
Deep Infra logo

Model details

DeepSeek V3 0324

DeepSeek V3 0324 is a recognized checkpoint within the DeepSeek model family, positioned as a refined update to the original DeepSeek V3 release and preceding later iterations in the lineage. The model sits alongside DeepSeek R1 0528 in the V3-era lineup and is treated by deployment-oriented guides as its own distinct variant rather than a re-tag of the base V3. Its naming convention marks it as a March 2024 release that continues the V3 line rather than initiating a new architecture family.

Community discussion around the model has centered on practical availability through cloud inference endpoints, where users have sought access to it on platforms supporting open-weight checkpoints. The 0324 refresh was noted by developers as a meaningful improvement over the earlier V3 release, particularly for stability in tool-calling workflows. As an open-weights release in the broader V3 lineage, it fits use cases where teams want a strong general-purpose text model with room for self-hosting and fine-tuning.

Deep Infradeepseek-ai/DeepSeek-V3-0324deepseekdeprecated

Quick Info

Powered by
Provider
Deep Infra
Model key
deepseek-ai/DeepSeek-V3-0324
Release date
Mar 24, 2025
Last updated
Mar 24, 2025
Input modalities
Output modalities
Capabilities

Cost

Input token cost
$0.24
Output token cost
$0.90

Limits

Output tokens
163,840 tokens
Context window
163,840 tokens

Transparent token rates

Compare DeepSeek V3 0324 pricing

Rates are shown per one million tokens. Combined means one million input plus one million output tokens.

Browse this family

Latest news about DeepSeek V3 0324

Deep Infra

CoverageBenchmark

DeepSeek released DeepSeek V3-0324 on March 24, 2025 as an update to its V3 model, with improvements in reasoning, coding, and frontend development capabilities according to Investing.com's reporting on the launch. The model is open source under the MIT license and is available on platforms including Hugging Face and O The V3-0324 update posted measurable gains on benchmarks including MMLU-Pro and LiveCodeBench, and became the highest-scoring non-reasoning model on Artificial Analysis' Intelligence Index at release — the first time an open-weights model led that category, surpassing proprietary non-reasoning models including Gemini 2

Videos about DeepSeek V3 0324

More models around DeepSeek V3 0324