ai.canshu.com
中文
VRAM Calculator Advisor & TCO API Pricing Compare Models Leaderboard GPUs

Search models, GPUs, and tools

↑↓ Navigate · Enter Open · ★ Watchlist prioritized ai.canshu.com
DeepSeek
Open-Weight DeepSeek · MoE 284B / A13B · Text · FP4 + FP8 Mixed

DeepSeek-V4-Flash (284B-A13B MoE) Architecture Parameters, Benchmarks & VRAM

DeepSeek-V4 efficient MoE (284B total / 13B active) with 1M context window and compressed sparse/MLA attention.

🧬 DeepSeek-V4-Flash (284B-A13B MoE) Architecture & Tensor Specifications

Verified directly against HuggingFace (deepseek-ai/DeepSeek-V4-Flash) config.json & Model Card

HF Class: DeepseekV4ForCausalLM

1. Backbone & Dimensions

deepseek_v4
Total / Active Params
284B / 13B
Hidden Size (d_model)
4,096
MoE Expert Dim
2,048
MoE Experts
256 Routed · Top-6 + 1 Shared
Activation / Norm
SiLU (SwiGLU) · ε=1e-6
Native Tensor Dtype
FP4 + FP8 Mixed

2. Attention & Layer Layout

MLA
Num Hidden Layers
43 Layers
Hidden Layer Layout
43-Layer Hybrid CSA + HCA Compressed Sparse Attention MoE + mHC
Attention Heads (Q / KV)
64 Q / 1 KV (64:1 GQA)
Head Dimension
512
Linear / KV Compression
CSA + HCA Hybrid Attention: uses only 10% KV cache vs DeepSeek-V3.2 at 1M context
Vision Encoder / MTP
MTP 1L

3. Context, RoPE & Tokenizer

Tokenizer
Native Context Length
1,048,576 tokens
RoPE Configuration
Decoupled RoPE (θ=10,000 · qk_rope_dim=64 · 原生 1M Context)
Vocab Size
129,280 (Untied)
Special Token IDs
BOS: 0 · EOS: 1

📋 DeepSeek-V4-Flash (284B-A13B MoE) Official API & System Card Specifications

Verified against official DeepSeek developer documentation & pricing

API Model ID: deepseek-v4-flash

1. Identity & Capabilities

API Spec
API Model ID
deepseek-v4-flash
Release Date
2026-07-31
Weight License
Open-Weight (API Spec)
Native Reasoning
✅ Supported
Function Calling & JSON Schema
✅ 原生支持
Batch API 异步批处理
✅ Supported (50% off)

2. Context, Output & Modalities

Limits
Max Context Window
1,024,000 tokens (1024K)
Max Output Tokens
384,000 tokens
Input Modalities
Text
Output Modalities
Text
Direct China Access
✅ Direct CN + RMB
Deployment Mode
Open-Weight + Cloud API

3. Official Pricing & Cache

CNY / 1M
Input / 1M Tokens
¥ 3
Cached Input / 1M
¥ 0.1 -97%
Output / 1M Tokens
¥ 9
Tiered Pricing Notes
Volcengine Ark: Volcengine Ark high-concurrency CN endpoint

🏆 Official Benchmark Results

Sourced from official System Card / Model Card & technical reports

Arena Elo ~1468 AA Index 47
📚 Knowledge High
MMLU-Pro 86.4 %

A more robust and challenging multi-task language understanding benchmark that extends MMLU by expanding multiple-choice options from 4 to 10, eliminating trivial questions, and focusing on reasoning-intensive tasks.

📚 Knowledge Max
Chinese-SimpleQA 78.9 %

Chinese short-form factual accuracy benchmark

🧠 Reasoning Max
GPQA Diamond 88.1 %

PhD-level Google-proof science reasoning questions

🧠 Reasoning Max
HMMT 2026 Feb 94.8 %
🧠 Reasoning Max
HLE 34.8 %

Humanity's Last Exam frontier academic benchmark

💻 Coding Max
SWE-bench Verified 79 %

A verified subset of 500 software engineering problems from real GitHub issues, validated by human annotators for evaluating language models' ability to resolve real-world coding issues by generating patches for Python codebases.

💻 Coding Max
LiveCodeBench 91.6 %

LiveCodeBench is a holistic and contamination-free evaluation benchmark for large language models for code.

🤖 Agentic Max
Terminal Bench 2.0 56.9 %
🤖 Agentic Max
MCPAtlas 69 %

Scaled Model Context Protocol multi-tool coordination

🎛️ Official Sampling & Reasoning Parameters

Recommended generation settings & API parameter constraints

Local Deployment & Think Max Mode
temperature=1 top_p=1 max_tokens=65,536

💡 Set temperature=1.0, top_p=1.0; allocate at least 384K context window for Think Max mode.

📊 DeepSeek-V4-Flash (284B-A13B MoE) Full Benchmark Scorecard

19 public benchmarks · ranks among 154 models · synced 2026-11-20

⚔️ Compare head-to-head

Source: ZeroEval / llm-stats.com (✓ = third-party verified, others vendor-reported).

⚡ DeepSeek-V4-Flash (284B-A13B MoE) Interactive VRAM & Token Speed Calculator

Concurrency (Batch):
MoE 284B / A13B·43L (d=4096)·64Q/1KV·Ctx 1M
Runnable Models →
512 GB Unified·1200 GB/s·Safe Usable 460.8 GB
Smooth Fit176.19 / 460.8 GB~99.7 tok/s
3. Weight Quantization
4. Per-Request Context16K (16,384 tok)
Verified: 2026-11-20·Roofline bandwidth/compute estimate (Batch=1)
Smooth Fit
176.19/ 460.8 GB usable

Physical 512 GB · Load 38.2%

Theoretical Decode Speed
~99.7tokens/s

MoE: reads only 7.88 GB active weights/step

Max Context at Batch=1
1M/ 1M

MLA compresses multi-batch KV footprint

Weights: 172.18 GBKV Cache (1×16K): 0.12 GBOverhead: 3.89 GB
Reserves ~10% unified memory for OS
🚀 Blazing Fast (Instant CoT)~5s for a 500-token response
Reference Launch CLI:
OLLAMA_CONTEXT_LENGTH=16384 ollama run deepseek-v4:flash

Mac Studio M5 Ultra 512G Fits DeepSeek-V4-Flash (284B-A13B MoE) · Cost & TCO

Approx. $9,499 (~$19/GB VRAM) · 190W TDP

📊 DeepSeek-V4-Flash (284B-A13B MoE) Quantization VRAM Breakdown (16K Context)

Quantization Bits/Weight Weights KV Cache (16K) Total VRAM Quality Note
FP16 / BF16 16 bpw 568 GB 0.12 GB 579.93 GB Lossless Full Precision
FP8 / Q8_0 8.5 bpw 301.75 GB 0.12 GB 308.35 GB 99.9% Near-Lossless
GGUF Q6_K 6.56 bpw 232.88 GB 0.12 GB 238.11 GB Ultra High Quality
GGUF Q5_K_M 5.69 bpw 202 GB 0.12 GB 206.6 GB High Balanced Quality
Q4_K_M / AWQ 4.85 bpw 172.18 GB 0.12 GB 176.19 GB Sweet Spot (Recommended)
GGUF IQ4_XS 4.25 bpw 150.88 GB 0.12 GB 154.46 GB Compact 4-bit
GGUF Q3_K_M 3.91 bpw 138.81 GB 0.12 GB 142.15 GB Moderate Perplexity Loss
GGUF Q2_K 3.15 bpw 111.83 GB 0.12 GB 114.63 GB Extreme Compression

🖥️ GPU & Apple Silicon Compatibility for DeepSeek-V4-Flash (284B-A13B MoE) (Q4_K_M)

Frequently Asked Questions

What are the exact architecture parameters of DeepSeek-V4-Flash (284B-A13B MoE)?

DeepSeek-V4-Flash (284B-A13B MoE) uses DeepseekV4ForCausalLM with 43 layers (43-Layer Hybrid CSA + HCA Compressed Sparse Attention MoE + mHC), hidden_size=4096, 64 Q heads / 1 KV heads (head_dim=512), and vocab_size=129,280.

How much VRAM does DeepSeek-V4-Flash (284B-A13B MoE) require?

At 4-bit quantization (Q4_K_M, 16K context), DeepSeek-V4-Flash (284B-A13B MoE) requires 176.19 GB VRAM. Full FP16 requires 579.93 GB.