DeepSeekEco TierMoE SparseZero Data Retention

V4 Pro 0813

Name: ARMES
Availability: InStock
Author: ARMES

Run V4 Pro 0813

Included on ECO plan

Context Window

1M (384K max output)

Input Token Price

$1.32

per 1,000,000 tokens

Output Token Price

$3.96

per 1,000,000 tokens

Speed Rating

Moderate

Architectural Profile & Capabilities

Terminal-Bench 2.1: 87.9%, DeepSWE: 62.7%, CyberGym: 83.3%, AutomationBench: 31.8%, Toolathlon-Verified: 74.1%, HLE w/ tools: 60.0%, GPQA Diamond: 90.1%, Artificial Analysis Intelligence Index: 53.2, AA Coding Index: 68.8, AA Agentic Index: 49.6 | The official GA release of V4 Pro, superseding the April preview (released Aug 13, 2026). Massive agentic leaps: Terminal-Bench 2.1 +15.8 pts, DeepSWE +49.9 pts, CyberGym +30.6 pts over preview. Broadly competitive with Opus 4.8 and Kimi K3 on agent benchmarks. Three reasoning modes: non-think, high, max. MIT licensed open weights. **Best for:** complex agentic coding, cybersecurity research, full-stack development, multi-step tool orchestration. **Caveats: benchmark scores are DeepSeek's own runs using unreleased DeepSeek Harness — Vals AI independently measured Terminal-Bench at 54.68% under the reference harness (33-pt gap). Pricing shifted to peak/off-peak on Aug 16, 2026. No system card published.**

Architecture Type: Mixture of Experts (MoE)

Sparse Mixture-of-Experts routing: activates only a subset of parameter experts per token, delivering frontier-level intelligence with exceptional efficiency.

Recommended Workloads & Primary Use Cases

complex agentic coding

cybersecurity research

full-stack development

ARMES Zero Data Retention

Calls to V4 Pro 0813 are routed through strict Zero Data Retention inference channels. Prompts and outputs are never stored, indexed, or monitored by DeepSeek.

View Privacy Architecture

More Models from DeepSeek

Compare other engines available in this laboratory.

View Roster (6)

V4 ProEco Tier

V4 Pro is a high-performance foundation model engineered by DeepSeek, accessible with Zero Data Retention on ARMES.

128K ctx$1.74/M

V4 Flash-0731Free Tier

Terminal-Bench 2.1: 82.7%, CyberGym: 76.7%, Toolathlon-Verified: 70.3%, DSBench-FullStack: 68.7%, DSBench-Hard: 59.6%, DeepSWE: 54.4%, NL2Repo: 54.2%, Agents' Last Exam: 25.2%, AutomationBench Public: 25.1%, Artificial Analysis Intelligence Index: 50 | The official GA release of V4 Flash, superseding the April preview (released Jul 31, 2026).

1M ctx$0.14/M

V4 FlashFree Tier

Strong on long-document and routine coding | Cost-effective high-volume tasks, long-document processing, routine coding, local deployment on consumer hardware.

1M ctx$0.14/M

Open App

Back to DeepSeek models roster

DeepSeekEco TierMoE SparseZero Data Retention

V4 Pro 0813

Run V4 Pro 0813

Included on ECO plan

Context Window

1M (384K max output)

Input Token Price

$1.32

per 1,000,000 tokens

Output Token Price

$3.96

per 1,000,000 tokens

Speed Rating

Moderate

Architectural Profile & Capabilities

Architecture Type: Mixture of Experts (MoE)

Sparse Mixture-of-Experts routing: activates only a subset of parameter experts per token, delivering frontier-level intelligence with exceptional efficiency.

Recommended Workloads & Primary Use Cases

complex agentic coding

cybersecurity research

full-stack development

ARMES Zero Data Retention

Calls to V4 Pro 0813 are routed through strict Zero Data Retention inference channels. Prompts and outputs are never stored, indexed, or monitored by DeepSeek.

View Privacy Architecture

More Models from DeepSeek

Compare other engines available in this laboratory.

View Roster (6)

V4 ProEco Tier

V4 Pro is a high-performance foundation model engineered by DeepSeek, accessible with Zero Data Retention on ARMES.

128K ctx$1.74/M

V4 Flash-0731Free Tier

1M ctx$0.14/M

V4 FlashFree Tier

Strong on long-document and routine coding | Cost-effective high-volume tasks, long-document processing, routine coding, local deployment on consumer hardware.

1M ctx$0.14/M