LLM Reference

DeepSeek V3 0324

Released
2025-03-24
Last refreshed
2026-06-29
Status
Researched 43d ago
Open sourceCommercial use: permittedCodingAgentsLong context

DeepSeek V3 0324 is worth evaluating for coding, agents, and long context when its provider route and context window match the workload.

Use it for

  • Teams evaluating coding, agents, and long context
  • Workloads that can use a 160k context window
  • Buyers comparing 3 tracked provider routes

Do not use it for

  • Vision or document-understanding workloads
  • Strict JSON or tool-calling flows
Specifications
Released
2025-03-24
Context
160k
Parameters
671B
Architecture
Decoder Only
Specialization
general
Openness
Open source
License
MITOSI-approvedCommercial use: permitted
Training
Pretrained
Created by

Advancing artificial general intelligence (AGI).

Hangzhou, Zhejiang, China
Founded 2023
Website
Pricing
Output / 1M
$1.12
Input / 1M
$0.270

Cheapest of 3 routes · Novita AI

About

DeepSeek: DeepSeek V3 0324 available via OpenRouter. Pricing: $0.2/1M input, $0.77/1M output.

DeepSeek V3 0324 is an open-source model in the DeepSeek V3 family. The structured metadata tracks a 160k-token context window. This page tracks provider routes through Fireworks AI, Microsoft Foundry, and Novita AI, with the cheapest tracked route listed at $0.27 input and $1.12 output per 1M tokens. Headline tracked benchmarks include HumanEval 85.5, Aider Polyglot 55.1, and Google-Proof Q&A 87.6.

Top use-case fit: coding, agents, and build tasks

Coding

Q/$ C

3 relevant benchmarks in the decision map.

Agents

Q/$ B

1 relevant benchmark in the decision map.

Long context

Included by capability and metadata signals in the decision map.

Provider price ladder

Compare all 3

Compare API pricing across 3 providers for input and output tokens, batch, and cached reads when available.

ProviderInput / 1MOutput / 1MRoute
Novita AI$0.270$1.12
Serverless
Fireworks AI$0.560$1.68
Serverless
Microsoft Foundry--
ServerlessProvisionedPartial

Available via routers & gateways(6)

Capabilities

No model capability flags are currently sourced.

Benchmark peer barsfor Coding

Benchmark scores(4)

Scores are benchmark-specific and are direction-aware: the same numeric gap can mean very different outcomes across suites. Use the leaderboard context and this model's provider route to decide whether the winning margin is meaningful for your workload.
BenchmarkScoreVersionSource
HumanEval85.52025-03https://arxiv.org/abs/2412.19437
Aider Polyglot55.12026-04https://aider.chat/docs/leaderboards
Google-Proof Q&A87.6https://github.com/deepseek-ai/DeepSeek-V3/blob/main/DeepSeek_V3.pdf
SWE-bench Verified76.8https://www.swebench.com/

Migration checks

No linked migration route is available for this model yet.

Frequently asked questions

What is the context window of DeepSeek V3 0324?

DeepSeek V3 0324 has a context window of 160k tokens.

How much does DeepSeek V3 0324 cost?

DeepSeek V3 0324 pricing ranges from $0.27/1M to $0.56/1M input tokens depending on the provider.

When was DeepSeek V3 0324 released?

DeepSeek V3 0324 was released on 2025-03-24.

Which providers offer DeepSeek V3 0324?

DeepSeek V3 0324 is available from 3 providers: Fireworks AI, Microsoft Foundry, Novita AI.

What benchmarks has DeepSeek V3 0324 been tested on?

DeepSeek V3 0324 has been evaluated on 4 benchmarks, including HumanEval, Aider Polyglot, Google-Proof Q&A, SWE-bench Verified.