Swift-1.5-Qwen3.8-Flash-Next requirements — can you run it?
· 27 quantizations measured · updated 2026-09-24 · source: ukisai/Swift-1.5-Qwen3.8-Flash-Next-GGUF
# llama.cpp (auto-downloads the GGUF) llama-cli -hf ukisai/Swift-1.5-Qwen3.8-Flash-Next-GGUF:IQ1_S # Ollama (pulls straight from Hugging Face) ollama run hf.co/ukisai/Swift-1.5-Qwen3.8-Flash-Next-GGUF:IQ1_S # LM Studio: search "ukisai/Swift-1.5-Qwen3.8-Flash-Next-GGUF" in the model browser
Commands load the exact quant measured on this page.
Every quantization: real file size vs what you actually need
| Quant | File (measured) | VRAM/RAM needed (8K ctx) | Runs on (first fit) |
|---|---|---|---|
| IQ1_S | 70.1 GB | 85.6 GB | RTX PRO 6000 96GB |
| IQ1_M | 72.0 GB | 87.9 GB | RTX PRO 6000 96GB |
| IQ2_XXS | 75.2 GB | 91.7 GB | Mac 128GB unified |
| IQ2_XS | 77.7 GB | 94.8 GB | Mac 128GB unified |
| IQ2_S | 77.8 GB | 94.9 GB | Mac 128GB unified |
| IQ2_M | 80.4 GB | 97.9 GB | Mac 128GB unified |
| Q2_K | 80.9 GB | 98.6 GB | Mac 128GB unified |
| IQ3_XXS | 88.0 GB | 107.1 GB | Mac 128GB unified |
| Q3_K_S | 89.4 GB | 108.8 GB | Mac 128GB unified |
| IQ3_XS | 91.9 GB | 111.8 GB | Mac 128GB unified |
| Q3_K_M | 92.0 GB | 111.8 GB | Mac 128GB unified |
| Q3_K_L | 93.2 GB | 113.3 GB | Mac 128GB unified |
| IQ3_M | 93.2 GB | 113.4 GB | Mac 128GB unified |
| IQ4_XS | 97.7 GB | 118.7 GB | Mac 256GB unified |
| IQ4_NL | 100.3 GB | 121.9 GB | Mac 256GB unified |
| Q4_0 | 100.6 GB | 122.2 GB | Mac 256GB unified |
| Q2_K_L | 107.1 GB | 130.1 GB | Mac 256GB unified |
| Q4_1 | 111.3 GB | 135.0 GB | Mac 256GB unified |
| Q4_K_S | 113.1 GB | 137.3 GB | Mac 256GB unified |
| Q3_K_XL | 119.3 GB | 144.7 GB | Mac 256GB unified |
| Q4_K_M | 119.6 GB | 145.0 GB | Mac 256GB unified |
| Q5_K_S | 128.1 GB | 155.2 GB | Mac 256GB unified |
| Q5_K_M | 134.7 GB | 163.1 GB | Mac 256GB unified |
| Q4_K_L | 139.3 GB | 168.6 GB | Mac 256GB unified |
| Q5_K_L | 151.1 GB | 182.8 GB | Mac 256GB unified |
| Q6_K | 168.1 GB | 203.2 GB | Mac 256GB unified |
| Q8_0 | 188.3 GB | 227.4 GB | Mac 256GB unified |
KV cache uses a flat +20% context/overhead rule because this repo does not publish its config; exact numbers may differ for long contexts. MoE models keep all experts in memory — total size counts, not just active params. Longer context grows the KV cache — use the selector above.
Can I run Swift-1.5-Qwen3.8-Flash-Next on my GPU? (IQ1_S recommended quant)
| Hardware | Memory | Verdict | Est. speed |
|---|---|---|---|
| RTX 3060 12GB | 12 GB | no | |
| RTX 4060 Ti 16GB | 16 GB | no | |
| RTX 3090 24GB | 24 GB | no | |
| RTX 4090 24GB | 24 GB | no | |
| RTX 5090 32GB | 32 GB | no | |
| RTX PRO 6000 96GB | 96 GB | RUNS | ~19 tok/s |
| Mac 16GB unified | 16 GB | no | |
| Mac 32GB unified | 32 GB | no | |
| Mac 64GB unified | 64 GB | no | |
| Mac 128GB unified | 128 GB | RUNS | ~4 tok/s |
| Mac 256GB unified | 256 GB | RUNS | ~6 tok/s |
| Mac 512GB unified | 512 GB | RUNS | ~9 tok/s |
| 32GB system RAM (CPU) | 32 GB | no | |
| 64GB system RAM (CPU) | 64 GB | no | |
| 128GB system RAM (CPU) | 128 GB | RUNS | ~0.9 tok/s |
| 256GB system RAM (CPU) | 256 GB | RUNS | ~0.9 tok/s |
| 1TB server RAM (CPU) | 1024 GB | RUNS | ~2 tok/s |
| 2TB server RAM (CPU) | 2048 GB | RUNS | ~2 tok/s |
Speed = decode tok/s estimated from memory bandwidth ÷ measured file size (× MoE active share). Real numbers vary ±30% by runtime and settings.
How much VRAM does Swift-1.5-Qwen3.8-Flash-Next need?
The honest answer: 85.6 GB at the IQ1_S quant with 8K context — that is the measured
file size (70.1 GB) plus KV cache and runtime overhead. The smallest published build needs
85.6 GB; the lossless one needs 227.4 GB.
Cheapest hardware that runs it: Mac Studio Ultra 256GB (~$6000, used market) — see which GPU to buy.
Can I run it with CPU offload?
Yes, if combined RAM+VRAM ≥ file size — but generation speed is limited by memory bandwidth. Expect roughly: DDR5 dual channel ~10-30 tok/s for small MoE actives, single digits for big ones, 0.1-1 tok/s when experts page from disk. CPU-only is fine for batch jobs, painful for chat.
Why do other sites show different numbers?
Most pages compute weights from a formula (params × bits/8). We use the actual file sizes published in GGUF repos, which include the embedding table, unquantized tensors, and container overhead — that is why our numbers can differ from a naive calculation.
FAQ
How much VRAM does Swift-1.5-Qwen3.8-Flash-Next need?
The measured answer: 85.6 GB at the IQ1_S quant with 8K context. The smallest published build needs 85.6 GB; the lossless (F16/BF16) build needs 227.4 GB. These are real GGUF file sizes from Hugging Face, not formula estimates.
Can I run Swift-1.5-Qwen3.8-Flash-Next on a 24GB GPU (RTX 3090/4090)?
Not at IQ1_S (85.6 GB). Nothing fits 24GB — the smallest build needs 85.6 GB. Use the hosted API or a smaller model.
How fast will Swift-1.5-Qwen3.8-Flash-Next run?
Decode speed is memory-bandwidth-bound. At IQ1_S: roughly 11 tok/s on an RTX 4090, 10 tok/s on an RTX 3090, 4 tok/s on an M-series Mac with 128GB, and 0.5 tok/s on CPU with dual-channel DDR4 — estimates from measured file size, MoE active-parameter share and memory bandwidth.
Can I run Swift-1.5-Qwen3.8-Flash-Next on CPU with system RAM?
Yes, if your usable RAM is at least the file size. At IQ1_S you need ~85.6 GB of RAM. CPU generation is memory-bandwidth-bound: expect single-digit tok/s for large MoE models, 10-30 tok/s for small active params, and below 1 tok/s when experts page from disk.
Why do VRAM numbers for Swift-1.5-Qwen3.8-Flash-Next differ between sites?
Most sites compute weights from a formula (params x bits/8). ModelFit uses the actual GGUF file sizes published on Hugging Face, which include the embedding table, unquantized tensors, and container overhead — so our numbers reflect what you really download and load.