EDGE OSOPEN-WEIGHT LLM COMPARISON
DeepSeek-V4-Flash vs MiniMax-M3Specs, VRAM and the hardware to run them
Both have 1M-token contexts and mixture-of-experts designs. DeepSeek-V4-Flash has 284B parameters with 13B active and an MIT license; MiniMax-M3 has 428B with about 23B active and adds image and video input, but its community license requires an agreement with MiniMax for commercial use.
Parameters
284B vs 427B
Active per token
13B vs 23B
Context
1M vs 1M
Memory, 1 user
177 GB vs 277 GB
Which should you choose?
CHOOSE DEEPSEEK-V4-FLASH IF
You want a permissive MIT license for commercial use and the smaller memory footprint.
CHOOSE MINIMAX-M3 IF
You need image and video input and can agree commercial terms with MiniMax.
Still deciding between DeepSeek-V4-Flash and MiniMax-M3?
Tell us your models, users and data. We will recommend a setup and run it inside your own building with Vanik Edge OS. Free, no obligation.
DeepSeek-V4-Flash vs MiniMax-M3 specs
From each developer’s model card. Green marks the lighter figure where less is better for running cost.
| Specification | DeepSeek-V4-Flash | MiniMax-M3 |
|---|---|---|
| Developer | DeepSeek | MiniMax |
| Total parameters | 284B | 427B |
| Active per token | 13B | 23B |
| Design | Mixture of experts | Mixture of experts |
| Experts | Mixture of experts (count not stated) | 128 routed experts + 1 shared, 4 routed active per token |
| Context window | 1M tokens | 1M tokens |
| Input and output | Text input, text output | Text, image and video input, text output |
| Layers | Not stated | 60 layers |
| Attention | Hybrid Compressed Sparse Attention and Heavily Compressed Attention | MiniMax Sparse Attention (attends to selected blocks of past tokens) |
| License | MIT | MiniMax Community License: commercial use needs a separate agreement with MiniMax |
| Released weight formats | FP4 + FP8 (native) | FP8 (official), BF16, 4-bit |
| Weights at FP4 + FP8 / 4-bit | About 168 GB | About 260 GB |
VRAM needed to run them
GPU memory at the lowest released precision. Each user needs their own KV cache, so memory grows with users and context length.
| GPU memory needed (FP4 + FP8 / 4-bit) | DeepSeek-V4-Flash | MiniMax-M3 |
|---|---|---|
| 1 user, 8K context | About 177 GB | About 277 GB |
| 1 user, 128K context | About 178 GB | About 307 GB |
| 16 users, 8K context each | About 178 GB | About 307 GB |
| 64 users, 8K context each | About 183 GB | About 403 GB |
Which GPUs can run DeepSeek-V4-Flash and MiniMax-M3
Smallest setup on each system, with 8K context per user. Green fits on one GPU or system, amber needs several in one server, grey needs more than one 8-GPU server.
| Hardware | DeepSeek-V4-Flash: 1 user | DeepSeek-V4-Flash: 16 users | MiniMax-M3: 1 user | MiniMax-M3: 16 users |
|---|---|---|---|---|
| NVIDIA DGX Spark 128 GB per system | 2 systems | 2 systems | No | No |
| NVIDIA L40S 48 GB per GPU | 8 GPUs | 8 GPUs | 8 GPUs | 8 GPUs |
| NVIDIA RTX PRO 6000 Blackwell 96 GB per GPU | 4 GPUs | 4 GPUs | 4 GPUs | 4 GPUs |
| NVIDIA H100 NVL 94 GB per GPU | 4 GPUs | 4 GPUs | 4 GPUs | 4 GPUs |
| NVIDIA H200 NVL 141 GB per GPU | 2 GPUs | 2 GPUs | 4 GPUs | 4 GPUs |
| AMD Instinct MI300X 192 GB per GPU | 2 GPUs | 2 GPUs | 2 GPUs | 2 GPUs |
| AMD Instinct MI325X 256 GB per GPU | Yes, 1 GPU | Yes, 1 GPU | 2 GPUs | 2 GPUs |
| NVIDIA DGX B200 1.44 TB per system | Yes, 1 system | Yes, 1 system | Yes, 1 system | Yes, 1 system |
Setups in amber or grey need several GPUs or servers working as one. Buying the GPUs is the easy part. Vanik Edge OS splits the model across them, keeps it running if a server fails and manages it all from one console. See how Edge OS clusters GPU servers
Which is faster?
On the same hardware, DeepSeek-V4-Flash should generate tokens about 1.8× faster for each user: it reads about 7.7 GB of weights per token (13B active at FP4 + FP8) versus 14 GB for MiniMax-M3. Token generation is limited mostly by memory bandwidth, so fewer active bytes means faster replies. Real throughput also depends on batching, the inference engine and the GPU.
FROM SIZING TO RUNNING
Sizing the hardware is the easy part.
Getting several GPUs to work as one system for a whole company is where most projects slow down. That is the part Vanik Edge OS does for you.
STEP 01 · THIS TOOL
Size it
Work out the GPU memory your models need and which GPUs or servers can hold them.
STEP 02 · VANIK EDGE OS
Cluster it
Edge OS joins your GPU servers into one cluster, splits a large model across them and keeps it running if a server fails. NVIDIA and AMD servers work side by side.
STEP 03 · VANIK EDGE OS
Run it
Your team gets AI apps with single sign-on and an audit trail, managed from one console, with your data staying inside your building.
DeepSeek-V4-Flash vs MiniMax-M3 questions
What is the difference between DeepSeek-V4-Flash and MiniMax-M3?
Both have 1M-token contexts and mixture-of-experts designs. DeepSeek-V4-Flash has 284B parameters with 13B active and an MIT license; MiniMax-M3 has 428B with about 23B active and adds image and video input, but its community license requires an agreement with MiniMax for commercial use.
DeepSeek-V4-Flash vs MiniMax-M3: which is better to run on-premise?
Choose DeepSeek-V4-Flash if you want a permissive MIT license for commercial use and the smaller memory footprint. Choose MiniMax-M3 if you need image and video input and can agree commercial terms with MiniMax.
Which needs less VRAM, DeepSeek-V4-Flash or MiniMax-M3?
DeepSeek-V4-Flash. For one user with 8K context it needs about 177 GB, versus about 277 GB for MiniMax-M3. For 16 users it needs about 178 GB versus 307 GB.
Which is faster, DeepSeek-V4-Flash or MiniMax-M3?
On the same hardware, DeepSeek-V4-Flash should generate tokens about 1.8× faster for each user: it reads about 7.7 GB of weights per token (13B active at FP4 + FP8) versus 14 GB for MiniMax-M3. Token generation is limited mostly by memory bandwidth, so fewer active bytes means faster replies.
Can I run DeepSeek-V4-Flash locally?
Yes. At FP4 + FP8 it needs about 177 GB of GPU memory for one user with 8K context, which fits on 2 × NVIDIA DGX Spark. A team of 16 needs about 178 GB: 2 × NVIDIA DGX Spark.
Can I run MiniMax-M3 locally?
Yes. At 4-bit it needs about 277 GB of GPU memory for one user with 8K context, which fits on 4 × NVIDIA H100 NVL. A team of 16 needs about 307 GB: 4 × NVIDIA H100 NVL.
Can I use DeepSeek-V4-Flash and MiniMax-M3 commercially?
DeepSeek-V4-Flash is released under MIT. MiniMax-M3 is released under MiniMax Community License: commercial use needs a separate agreement with MiniMax. Check each license before a commercial deployment.
How are these numbers calculated?
Memory figures come from our VRAM calculator method: weights at the lowest released precision, plus the KV cache for each user’s context, plus about 5 to 10% runtime overhead, keeping 10% of each GPU free. Where a developer does not publish the cache size we estimate it from the architecture. These are planning estimates, not benchmarks.
Related comparisons
Sources: Hugging Face: deepseek-ai/DeepSeek-V4-Flash-0731; Hugging Face blog: DeepSeek-V4; Hugging Face: MiniMaxAI/MiniMax-M3. Memory figures are planning estimates; real use depends on the inference engine and settings. Last checked November 12, 2026.