DeepSeek V4

DeepSeek V4 Release & Features | 1.6T-Param MoE, 1M Context, Open-Source MIT

DeepSeek V4 is here: 1.6T-parameter MoE (49B active), 1M-token context, CSA+HCA hybrid attention, SWE-bench 80.6% | Released and open-sourced April 24, 2026

Last Update: April 2026

DeepSeek V4 was officially released and open-sourced under the MIT license on April 24, 2026, with weights on Hugging Face. It ships in two versions: V4-Pro (1.6T total / 49B active) for high-end reasoning and agentic coding, and V4-Flash (284B / 13B) for faster, lower-cost use. Both offer a 1M-token context window powered by a hybrid attention architecture (CSA + HCA) that cuts per-token compute to ~27% of V3.2 and KV-cache memory to ~10% at 1M context. V4-Pro scores 80.6% on SWE-bench Verified — the highest among open models, tied with Gemini 3.1 Pro (80.6%) and ahead of GPT-5.4 (77.2%). API pricing is $0.435/$0.87 per M tokens (Pro) and $0.14/$0.28 (Flash), roughly 5-30x cheaper than closed frontier models.

📅 Release Timeline

2024.12

DeepSeek-V3 Released

671B params, 37B active, MoE architecture

2025.01

MODEL1 Code Appears

MODEL1 identifier found in GitHub FlashMLA repo

2026.04

V4 Released & Open-Sourced

Launched April 24, 2026 (MIT). V4-Pro 1.6T/49B, V4-Flash 284B/13B, 1M context

2026.04

Enterprise Version Live

Atlas Cloud syncs V4 enterprise service on release

🚀 Core Features

From the official April 24, 2026 release

🌐

Two Versions: Pro & Flash

DeepSeek V4 ships in two open-source versions: V4-Pro for high-end reasoning and agentic coding, and V4-Flash for faster, lower-cost workloads. Both focus on text, code and reasoning.

  • V4-Pro: 1.6T total / 49B active parameters
  • V4-Flash: 284B total / 13B active parameters
  • Both support a 1M-token context window
  • Focused on text, code and reasoning
Source: DeepSeek Official
🏗️

1.6 Trillion Parameter MoE

V4-Pro features 1.6 trillion total parameters with only 49B active per token via Mixture-of-Experts. This delivers frontier performance at a fraction of closed-model cost.

  • V4-Pro: 1.6T total, 49B active per token
  • Mixture-of-Experts (MoE) architecture
  • API pricing: $0.435/$0.87 per M tokens (Pro)
  • Open-source under MIT license
Source: DeepSeek Official
📚

Million-Token Context

Both V4-Pro and V4-Flash support a 1M-token context window (max output ~384K), enough to process entire books, large codebases or ultra-long documents.

  • 1M-token context by default on both versions
  • Max output around 384K tokens
  • Process entire books (~500K words)
  • Analyze complete large project codebases
Source: DeepSeek Official
🧬

CSA + HCA Hybrid Attention

V4 combines Compressed Sparse Attention (CSA) and Heavily Compressed Attention (HCA) on top of its MoE design, making a 1M-token context window practical and cheap.

  • Hybrid of CSA and HCA attention
  • ~27% of V3.2 per-token compute at 1M context
  • ~10% of V3.2 KV-cache memory at 1M context
  • Enables low-cost ultra-long context
Source: DeepSeek Official

Ultra-Efficient 1M Context

Thanks to the CSA+HCA hybrid attention, V4 serves a full 1M-token context at a fraction of prior compute and memory cost — the core efficiency story of the release.

  • Per-token compute ~27% of V3.2 at 1M context
  • KV-cache memory ~10% of V3.2 at 1M context
  • Makes million-token workloads affordable
  • Strong throughput for agentic, long-context tasks
Source: DeepSeek Official
🧠

Strong Reasoning & Math

V4 posts strong reasoning and math results: GPQA Diamond 90.1%, MMLU-Pro 87.5%, GSM8K 92.6%, and Terminal-Bench 2.0 67.9% for agentic tasks.

  • GPQA Diamond: 90.1%
  • MMLU-Pro: 87.5%
  • GSM8K: 92.6%
  • Terminal-Bench 2.0: 67.9%
Source: DeepSeek Official Benchmarks
💰

Far Cheaper Than Closed Models

DeepSeek V4-Pro API pricing is $0.435/M input and $0.87/M output (after a 75% cut); V4-Flash is $0.14/$0.28. Open-source means free self-hosting.

  • V4-Pro: $0.435 input / $0.87 output per M tokens
  • V4-Flash: $0.14 input / $0.28 output per M tokens
  • Roughly 5-30x cheaper than closed frontier models
  • Open-source (MIT): free to self-host
Source: DeepSeek Official Pricing
🎯

Leads Open Models in Coding

V4-Pro scores 80.6% on SWE-bench Verified — the highest among open models, tied with Gemini 3.1 Pro (80.6%) and ahead of GPT-5.4 (77.2%).

  • SWE-bench Verified: 80.6% (highest open model)
  • LiveCodeBench Pass@1: 93.5
  • Codeforces rating: 3206
  • Strong multi-language and repo-level coding
Source: DeepSeek Official Benchmarks

🔬 Technical Deep Analysis

Inside the V4 hybrid attention architecture

Architecture Innovation

  • MoE backbone with two sizes (1.6T/49B Pro, 284B/13B Flash)
  • Hybrid attention combining CSA and HCA
  • 1M-token context window by default
  • Open-sourced under the MIT license

Efficiency at 1M Context

  • Per-token compute ~27% of V3.2
  • KV-cache memory ~10% of V3.2
  • CSA compresses sparse long-range attention
  • HCA heavily compresses retained context

Performance & Cost

  • SWE-bench Verified 80.6% (highest open model)
  • LiveCodeBench Pass@1 93.5, Codeforces 3206
  • V4-Pro pricing $0.435/$0.87 per M tokens
  • V4-Flash pricing $0.14/$0.28 per M tokens

📊 V3 vs V4 Comparison

Main upgrade points overview

Feature
V3
V4
Parameters
671B total / 37B active
Pro 1.6T/49B · Flash 284B/13B
Modality
Text only
Text, code, reasoning focused
Context
128K tokens
1M tokens (max output ~384K)
Memory
KV Cache
~10% of V3.2 KV-cache memory at 1M context
Attention
Standard MLA
CSA + HCA hybrid attention (~27% compute at 1M)
Reasoning
Standard
GPQA 90.1%, MMLU-Pro 87.5%, GSM8K 92.6%
API Price (input)
$0.28/1M tokens
$0.435/1M (Pro) · $0.14/1M (Flash)
Coding (SWE-bench)
~70%
80.6% (highest open model)
Open Source
Yes, Apache 2.0
Yes, MIT
Hardware
H800 optimized
Blackwell + Huawei Ascend + Cambricon

🏆 V4 vs Frontier Models

How DeepSeek V4 stacks up against GPT-5.4, Claude 4.6, and Gemini 3.1 Pro

Feature
DeepSeek V4
GPT-5.4
Claude 4.6
Gemini 3.1 Pro
Release Date
Apr 24, 2026
Mar 5, 2026
Feb 5, 2026
Feb 19, 2026
Context Window
1M
1.05M
1M
1M
Architecture
MoE + CSA/HCA
MoE
Dense
MoE
Input Price
$0.435/M (Pro)
$2.50/M
$5.00/M
$2.00/M
Output Price
$0.87/M (Pro)
$15.00/M
$25.00/M
$12.00/M
SWE-bench
80.6%
77.2%
80.8%
80.6%
LiveCodeBench
93.5 (Pass@1)
N/A
N/A
N/A
Multimodal
Text/code/reasoning focused
Text + Vision + Audio
Text + Vision
Native (text/image/video/audio)
Open Source
✅ MIT
❌ Closed
❌ Closed
❌ Closed
Local Deploy
✅ Free self-host
❌ API only
❌ API only
❌ API only

📎 Information Sources

The following is based on DeepSeek's official release (2026-04-24)

🟢

Official Release

  • Released and open-sourced (MIT) on April 24, 2026, weights on Hugging Face
  • V4-Pro 1.6T/49B and V4-Flash 284B/13B, both 1M-token context
  • CSA + HCA hybrid attention: ~27% compute, ~10% KV-cache memory vs V3.2 at 1M
🟢

Official Benchmarks

  • SWE-bench Verified 80.6% (highest open model, tied with Gemini 3.1 Pro)
  • LiveCodeBench Pass@1 93.5, Codeforces 3206
  • MMLU-Pro 87.5%, GPQA Diamond 90.1%, GSM8K 92.6%, Terminal-Bench 2.0 67.9%
🟢

Official Pricing & Access

  • V4-Pro $0.435/$0.87 per M tokens, V4-Flash $0.14/$0.28 per M tokens
  • Available via chat.deepseek.com (Expert/Instant Mode), official API, Atlas Cloud
  • Legacy deepseek-chat and deepseek-reasoner retire on July 24, 2026
⚠️ Disclaimer: The above reflects DeepSeek's official release on 2026-04-24. Some third-party benchmark figures may shift as evaluations are updated.

🎁 How to Use V4 Today

DeepSeek V4 is live on Atlas Cloud

✅ Available now, no waiting
✅ No server configuration needed, direct API calls
✅ Compatible with V3 code, zero upgrade cost
✅ Enterprise-grade stability and tech support
1

Register Atlas Cloud Now

Register an account and get free credits

2

Get Your API Key

Create an API key in the console

3

Switch Model

Set model to 'deepseek-v4-pro' (or 'deepseek-v4-flash') in your API request

📬 Subscribe to V4 Updates

Get DeepSeek V4 news, tutorials and updates

Release and update notificationsTechnical analysis articlesUsage tutorialsSpecial offers

DeepSeek V4 Is Live — Try It Now

Register Atlas Cloud now and start using V4 immediately

Register Now