Can I run DeepSeek V4.1 Flash 1017 on an RX 9060 XT 16GB?
Key facts
No: it needs about 446 GB and this card has 16 GB of graphics memory.
It needs about 445.5 GB at standard quality (estimate). Checked . mrfakename Q4_K_M GGUF (plain quant, no imatrix).
For this model we could not count the memory a conversation takes, so real use needs more than shown here.
Updated · DeepSeek V4.1 Flash 1017 on every card
How much memory it needs
That is about 28.9 times what this card has free.
Get a 1-line alert when a faster quant for DeepSeek V4.1 Flash 1017 drops — or when GPUs that run it drop >10%
Get a 1-line alert when a new low-bit GGUF/MLX quant fits 128–256 GB rigs or when multi-GPU & Mac Studio prices drop >10% — weekly, zero spam.
What runs DeepSeek V4.1 Flash 1017 instead
Nothing among the 43 graphics cards, PCs and Macs on our list runs it fully.
See our picks for DeepSeek V4.1 Flash 1017 in Find hardware. Add your budget to see cheaper machines.
For comparison, an RX 9060 XT 16GB costs $522€564, new (cheapest in-stock board on GPU PRIX's German board list). Card details
DeepSeek V4.1 Flash 1017 on other graphics cards
Doesn't fit10
Not enough graphics memory: it needs about 445.5 GB.
- RTX 3060 12GBhas 12 GB
- RTX 4060 Ti 16GBhas 16 GB
- RTX 3090has 24 GB
- RTX 4090has 24 GB
- RTX 5080has 16 GB
- RX 9060 XT 16GBhas 16 GB
- RTX 5090has 32 GB
- RX 7900 XTXhas 24 GB
- Dual RTX 3090has 48 GB
- Dual RTX 4090has 48 GB
More about the model itself, what it is good at and its licence: DeepSeek V4.1 Flash 1017.
The numbers behind this answer
A model comes in several sizes. Q4 is the standard one most people use; the bigger ones answer a little better but need more memory. Each row adds up the model file, the memory a conversation takes, and some working space, and compares the total with what is free on this card.
Swipe the table sideways to see every column.
| Quality | Conversation | Result | Model file | Conversation memory | Working space | Total needed | Free memory |
|---|---|---|---|---|---|---|---|
| Q4 · standard (smallest that keeps quality) | ~12 pages8,192 tokens | Doesn't fit | 444.7 GBmeasured file | 0.0 GBnot counted | 0.8 GB | 445.5 GB | 15.4 GB |
| Q5 · a bit sharper, a bit bigger | ~12 pages8,192 tokens | Doesn't fit | 541.9 GB | 0.0 GBnot counted | 0.8 GB | 542.7 GB | 15.4 GB |
| Q8 · near full quality, almost twice Q4 | ~12 pages8,192 tokens | Doesn't fit | 816.6 GB | 0.0 GBnot counted | 0.8 GB | 817.4 GB | 15.4 GB |
| FP16 · full quality, over three times Q4 | ~12 pages8,192 tokens | Doesn't fit | 1,526.4 GB | 0.0 GBnot counted | 0.8 GB | 1,527.2 GB | 15.4 GB |
| Q4 · standard, longer conversation | ~49 pages32,768 tokens | Doesn't fit | 444.7 GBmeasured file | 0.0 GBnot counted | 0.8 GB | 445.5 GB | 15.4 GB |
speed: not estimated for MoE yet. This model only uses part of itself for each word (a mixture of experts), and our speed estimate does not handle that yet.
Our estimate leaves out the conversation memory for this model: real use needs more than the figures shown.
Sources and method
Memory and speed on this page are our estimate, not a measured test. Checked . mrfakename Q4_K_M GGUF (plain quant, no imatrix).
Mainline llama.cpp does not support DeepSeek V4.1 yet. The mrfakename community GGUFs need open PR https://github.com/ggml-org/llama.cpp/pull/28696 (still open, checked 24 Oct 2026) and are text-only. Unsloth's GGUFs need llama.cpp PR https://github.com/ggml-org/llama.cpp/pull/30817 or Unsloth's prebuilt. llama.cpp PR #30817 was still a draft, checked 24 Oct 2026. A tester reported wrong tokens when llama-server reuses a cached prompt prefix (workaround: cache_prompt false), and maintainers asked to replace the MXFP8 tensors, so the UD-Q3_K_XL/Q4_K_XL/Q8_K_XL files may change (https://github.com/ggml-org/llama.cpp/pull/30817). That support is not mainline, and it is not PR #28696. Ollama lists deepseek-v4.1-flash:cloud and deepseek-v4.1-flash:1017-cloud (both cloud tags, no local tag), checked 24 Oct 2026 (https://ollama.com/library/deepseek-v4.1-flash). Official weights: FP8 with FP4 experts, 510.3 GB on HF. Language backbone stored as 551.9B; the architecture section says "552B backbone parameters" (1017 is a minor update of https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash). Weights for this row: https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017. Also 196.6B Engram conditional-memory parameters (sparse token lookup), 14.2B MTP head and 0.49B vision encoder. Total logical weights 763,205,315,794 (HF safetensors metadata, stored here as 763.21B). Q5, Q8 and FP16 use that total, not the 551.9B backbone: the Q4 file includes the Engram tables, and a backbone-only Q5 estimate was smaller than that file. Causal encoder-decoder, 40 layers (20 encoder + 20 decoder), 384 routed experts plus 1 shared, 6 routed per token, CSA2 compressed attention. KV is projected from encoder states with compressed and sliding attention (persistent KV about 1/8 of V4-Flash). The config sets compress_ratios and kv source layers 2, 8, 14 and 20, so we do not compute that cache and the conversation-memory figure stays at 0. Unsloth GGUF sizes, shards summed from the Hugging Face tree, decimal GB, checked 24 Oct 2026 (https://huggingface.co/unsloth/DeepSeek-V4.1-Flash-1017-GGUF, MIT): UD-IQ1_S 232.98 GB (232,979,608,475 bytes), UD-IQ1_M 250.40, UD-IQ2_XXS 300.48, UD-Q2_K_XL 334.79, UD-IQ3_XXS 394.74, UD-IQ3_S 423.63, UD-Q3_K_XL 485.07, UD-Q4_K_XL 500.71, UD-Q8_K_XL 508.38. The smallest llama.cpp-branch file on our list is UD-IQ1_S 232.98 GB. The mrfakename Q2_K file is 264.52 GB and is not the smallest. Each Unsloth quant includes an Engram lookup-table shard (the last shard): IQ1 110.60 GB (IQ4_NL), IQ2_XXS and Q2_K_XL 135.17 GB (Q5_K), IQ3 161.28 GB (Q6_K), Q3_K_XL, Q4_K_XL and Q8_K_XL 202.76 GB (MXFP8). Without that shard, UD-IQ1_S is 122.38 GB. Unsloth says a 1-bit quant needs 128–145 GB of RAM plus VRAM with the Engram table on SSD via mmap (https://unsloth.ai/docs/models/deepseek-v4.1-flash). That is a vendor claim, not tested by us. A 128 GB unified-memory machine is borderline on that claim. We do not score it as a fit: macOS leaves about 96 GB usable on a 128 GB Mac, and the full UD-IQ1_S file fits none of the 34 machines fully in VRAM. mmproj-F16 is 0.97 GB (971,004,064 bytes) and the DSpark drafter Q8_0 is 7.98 GB (7,976,914,240 bytes). Both are separate files and are not inside the quant totals. Other runtimes (not llama.cpp). Not a fit score. Baekpica VQ-GGUF [self-reported]: experimental low-bit release, routed experts about 1.6 bits per weight, files may be replaced, base file 113,556,639,424 bytes, runtime ds4-dfm-rs, serving speed not measured yet (https://huggingface.co/Baekpica/DeepSeek-V4.1-Flash-1017-VQ-GGUF). dsx 2-bit build v4b [self-reported]: 148 GB, KL 0.373, top-1 74.4%, perplexity 10.68. The card's 2× RTX 3090 plus 128 GB DDR4 figure is 28.9 tok/s decode on build v4, which it says has the same expert layout as v4b. RTX 5090 plus 128 GB DDR5 is 12.0 tok/s decode on build v2 (https://huggingface.co/lxe/DeepSeek-V4.1-Flash-1017-dsx-2bit). Project Maya / Maya-S, MIT [self-reported]: model shards 178.19 GB plus Engram Q6_K shards 161.28 GB, 339.48 GB together (https://huggingface.co/peasantsmith/DeepSeek-V4.1-Flash-1017-Maya-GGUF), runtime https://github.com/mw00/project-maya. [community] One post reports an RTX 3090 plus 64 GB RAM at about 4 tok/s decode. That post states 178.2 GB of Maya-S plus 202.8 GB of Engram, which is not this Q6_K Engram size (https://x.com/ItsmeAjayKV/status/2112515177954598912). DeepSeek's DeepInfer (https://github.com/deepseek-ai/DeepInfer, Apache-2.0) is a datacenter serving stack for NVIDIA SM100 GPUs and Huawei Ascend 950 NPUs, not a home-rig runtime.
Some downloads of this model (MLX) only run on Apple Silicon Macs, not on a graphics card.
Measured by others
Someone else ran DeepSeek V4.1 Flash 1017 and published the speed (community measurement, single run). They may have used a smaller or bigger version than the standard size above, so their result can differ from our answer: "How they measured" has the details.
- 2× RTX 3090, llama.cpp, UD-IQ1_M decode
How they measured
Community claim, not verified by us. One post on 20 Oct 2026 reports Unsloth UD-IQ1_M on 2× RTX 3090 with llama.cpp at 12.3 tokens per second decode and about 90 tokens per second tuned prefill. We did not run that test. The prefill figure is the post's, not a second decode measurement.
Source: lxe on X, published 20 Oct 2026
Questions
- Can I run DeepSeek V4.1 Flash 1017 on an RX 9060 XT 16GB?
- No: it needs about 446 GB and this card has 16 GB of graphics memory.
- How much memory does DeepSeek V4.1 Flash 1017 need?
- About 445.5 GB at the standard quality (Q4) for a conversation of up to about 12 pages of text (8,192 tokens), by our estimate: 444.7 GB for the model file (measured), 0.0 GB to hold the conversation (not modelled for this model, so real use needs more) and 0.8 GB of working space.
- How fast is DeepSeek V4.1 Flash 1017 on an RX 9060 XT 16GB?
- We don't give a speed for this model yet. This model only uses part of itself for each word (a mixture of experts), and our speed estimate does not handle that yet.
Speeds and memory are our estimate from the hardware specs, not a measured test. Prices are dated and may change. How we calculate