Inference
EN
DeepSeek

DeepSeek V3 0324

DeepSeek V3 0324 is a standard model from DeepSeek in the DeepSeek family. 16 benchmarks count toward its index, in 7 categories.

  • Open weights
  • Standard
  • Input: text
  • Released 2025-03
RankedEnough results for an index.
IndexAll categories, out of 100.37.1±5.0Median 51.0
CoverageIndex weight that has results.90%
SpeedOutput tokens per second.—Median 67/s
First tokenSeconds to the first output token.—Median 1.3 s
Input / 1MUS dollars per 1M input tokens.—Median $0.60
Output / 1MUS dollars per 1M output tokens.—Median $2.55
ContextMaximum tokens in one request.128KMedian 262K
EloLMArena rating and rank.1375±4 #184Median 1422

Run locally

684.53B parameters
16-bit
1438.5 GB
8-bit
764.7 GB
4-bit
436.7 GB
Weights
Hugging Face (opens in a new tab) ↗

Memory need with a short context. The 4-bit copy needs more than 128 GB. Compare local models

Capabilities

Index per category, out of 100.
Out of 100
AgenticMulti-step tasks with tools.2 counted34.0190 / 218
CodingCode writing and repair.4 counted40.9179 / 209
ReasoningLogic problems and puzzles.2 counted36.0183 / 227
MultimodalTasks with images and text.None counted——
KnowledgeFacts and expert knowledge.4 counted37.6210 / 268
MultilingualTasks in many languages.1 counted40.1—
InstructionTasks with strict prompt rules.1 counted27.3—
MathMath problems.2 counted38.0129 / 168

Results

16 counted
BenchmarkThe benchmark name.ResultThe score from the source.PlacePlace among results on this benchmark.
MATH 500Math88.6% ±1.430 / 59
AIMEMath52.2% ±1.760 / 95
τ²-Bench Tool-Agent-User EvaluationAgentic47.1%97 / 149
Artificial Analysis Omniscience AccuracyKnowledge24.3%108 / 164
MMLU ProKnowledge79.5% ±0.494 / 135
Artificial Analysis IFBenchInstruction41.0%102 / 146
Critical Physics TasksReasoning0.0%150 / 214
LiveCodeBenchCoding65.5% ±1.2100 / 142
SWE-bench Verified · SWE-bench/experimentsCoding42.0%47 / 65
Artificial Analysis GPQA DiamondKnowledge65.5%144 / 199
GDPval-AA normalizedAgentic0.0%101 / 133
Artificial Analysis Long Context ReasoningReasoning40.7%167 / 217
Artificial Analysis SciCodeCoding39.0%91 / 114
GPQA Diamond · Vals AIKnowledge61.6% ±2.4109 / 135
Artificial Analysis Humanity's Last ExamKnowledge4.7%183 / 222
MMLU-ProXMultilingual70.5%15 / 17
IOI v1Coding1.7% ±0.957 / 61
MGSMMultilingual91.7% ±1.726 / 74
TaxEval v2Finance71.1% ±0.983 / 143
CorpFin v2Finance54.7% ±1.092 / 131
MedQAHealthcare82.0% ±0.468 / 95
LegalBenchLegal77.7% ±0.4107 / 149

Sources

BenchLMCC BY-NC 4.0Vals AINo licence statedSWE-bench teamNo licence stated

History

5 releases
Index37.9 → 37.1
23 Oct 202612 Nov 2026

Questions

How good is DeepSeek V3 0324?
Its index is 37.1 out of 100. That is place 226 of 283 ranked models.
What is DeepSeek V3 0324 best at?
Math: 38.0 out of 100, place 129 of 168.
What is the context window of DeepSeek V3 0324?
128K tokens.
Is DeepSeek V3 0324 open source?
Yes. Its weights are published.
When was DeepSeek V3 0324 released?
March 2025.

Badge

DeepSeek V3 0324 on Inference360: place 226 of 283, index 37.1
HTML
<a href="https://inference360.ai/models/deepseek-v3-0324"><img src="https://inference360.ai/badges/deepseek-v3-0324.svg" alt="DeepSeek V3 0324 on Inference360: place 226 of 283, index 37.1" height="20"></a>
Markdown
[![DeepSeek V3 0324 on Inference360: place 226 of 283, index 37.1](https://inference360.ai/badges/deepseek-v3-0324.svg)](https://inference360.ai/models/deepseek-v3-0324)

Compare with

MiniCPM5-2B37.1Nemotron 3.5 Lightning 30B A3B NVFP437.0Granite 4.2 8B37.6Claude 3.7 Sonnet37.7DeepSeek-R138.9

More from DeepSeek

All models
R1 Distill Llama 70BUnrankedDeepSeek-R138.9DeepSeek R1 Distill Qwen 32BUnrankedDeepSeek-R1-052844.6DeepSeek V329.9