DeepSeek V4.1 Flash hands-on tests and comparisons

In a timed exploit challenge only frontier models finished: Claude Mythos 5.1 retrieved the flag in 3h28m for about $100 and GPT-5.6 Sol took 6h10m, while DeepSeek V4.1 Flash and Kimi K3.1 Pro both stalled at identifying the memory leak.

DeepSeek's API now redirects V4 Flash Vision Exp calls to V4.1 Flash 1017; a llama.cpp fix made with the model on a 192GB M2 Ultra was merged, logging 28.1 minutes at 215 tok/s prompt and 30.8 tok/s output, with the smallest GGUF at 98.6GB.

On a roughly $6k Strix Halo plus two RX 7900 XTX setup tuned via Project Maya, DeepSeek V4.1 Flash 1017 reaches 41.3 decode and 1,291 prefill tok/s, with 8.3x faster prefill on a 130K prompt.

First try of deepseek-v4.1-flash-1017 on a local rig hits about 100 tok/s with the Maya-S quant, with a caveat that quantization likely costs two or three benchmark points.

Six community recipes for DeepSeek V4.1 Flash ran on the same four DGX Sparks: all passed 71 of 75 qeval tasks and found every passphrase in a 1M-token prompt, while single-user speed ranged from 77 to 134 tok/s and vLLM served 32 concurrent users at 288 tok/s.

After an analysis blamed his setup and failed requests for TTFT numbers, the author argues the critique ran 148 requests with zero image input, making it benchmaxxed rather than a real multi-agent workflow.

On short prompts, MiniMax-M3.1-Flash led at 0.89s with DeepSeek V4.1 Flash at 0.91s, while GLM, Kimi and GPT-6.1-sol all exceeded 4s.

The author argues 1M context is overkill for most coding tasks and the sweet spot for DeepSeek V4.1 Flash is 300K–500K, recommending a 400K default with manual compaction since agents count max output toward the limit.

The author ran 11 open-weight models locally on a Mac mini; DeepSeek V4.1 Flash and GLM-5.3 Flash looked stable but drifted back toward Lincoln's original wording, suggesting memorization, while the largest models streamed from SSD and took ~36 of 45 hours.

Over 7.3 hours and 1,214 requests on a GB300, the model slept 51% of the time, background evals got 3.7 free GPU-hours, and 85% of requests waited nothing extra (1.3s average added wait).

Using the official API at high thinking and two turns totaling 16M tokens, the author had V4.1 Flash generate a three.js night scene with frogs, a bat and fish, calling the result high quality.

Using DeepSeek V4.1 Flash via Jarvislabs Token Factory inside opencode, the author finds changes appear almost instantly while iteratively building a small game, all recorded on the same VM.