Model Detail Human reviewed

DeepSeek V4 Flash 0731

DeepSeek V4 Flash 0731 is the official open-weight release that supersedes the preview model and adds the DSpark speculative decoding module. Its 304B-parameter, approximately 13B-active MoE checkpoint uses mixed FP4/FP8 weights, supports a 1M-token context window, three reasoning-effort levels, and agentic tool use. It is downloadable from Hugging Face and is also available through the Ollama Cloud tag deepseek-v4-flash:0731-cloud.

DeepSeekDEEPSEEK V4mit2026-07-31
Parameters
304B
13B active
Context window
1M
Long context
Architecture
csa hca mhc moe
moe
Quality score
96
Planner signal

Task Fit

Code AgentSupported

Tool use, repo work, terminal workflows, and coding benchmarks.

CodeSupported

Code generation, debugging, refactoring, and benchmark signal.

ChatSupported

General writing, Q&A, and assistant use.

RAGSupported

Document QA benefits from long context and instruction following.

VisionNot a fit

Not marked for vision in the current library.

Image GenerationNot a fit

Not marked for image generation in the current library.

Video GenerationNot a fit

Not marked for video generation in the current library.

VoiceNot a fit

Not marked for voice in the current library.

Source Confidence

Overallhigh · 100/100
ParametersReviewed / seeded
Task fitReviewed / seeded
MemorySeeded artifact
LicenseSource / seed
BenchmarksAvailable
Hardware fitCalculated

Variants and Quant Artifacts

Choose the artifact first; hardware fit follows from RAM, VRAM, format, and runtime.

1 artifacts
QuantFormatQualityMin RAMReco RAMRuntimeAction
MIXED_FP4_FP8safetensorsbalanced192GB256GBtransformers, vllm, sglang Plan with this

Benchmarks

Terminal Bench82.7%
Cybergym76.7%
Toolathlon Verified70.3%
Dsbench Fullstack68.7%
Dsbench Hard59.6%
Deep SWE54.4%
Nl2repo54.2%
Agents Last Exam25.2%
Automationbench Public25.1%

Source and Review

Hugging Facedeepseek-ai/DeepSeek-V4-Flash-0731
Ollama Clouddeepseek-v4-flash:0731-cloud
VerificationHuman reviewed
Artifact sourceofficial-deepseek-v4-flash-0731-weights
Default variantDeepSeek V4 Flash 0731 Reasoning
Tool callingSupported

Similar Models

Continue planning your local AI setup