Can you run IQuest Q1 locally?
IQuest family · 2026 · 320.3B parameters (15B activated per token) · Hugging Face model card
IQuest Q1 is a mixture-of-experts model from IQuest built for agentic coding, reasoning and multi-step tool use, with approximately 320B total parameters and an estimated 15B activated per token (vendor card: Total Parameters 320B; Activated Parameters 15B). The HF safetensors total is 320,318,615,552, consistent with the announced 320B. config.json declares 256 routed experts with 8 selected per token, 88 layers, hidden size 3072, 48 attention heads, 8 KV heads, head dim 128, vocab 160000, and a sliding window of 4096. Context is 524,288 tokens (config.json max_position_embeddings; card lists 524,288). Licensed under the custom IQuest-Q1 license (other). Local VRAM fit uses the 15B active count: roughly 9 GB at Q4-class quantization (calculated). Measured speeds: not yet published here.
Minimum: 10 GB+ memory (Q4-class on 15B active, calculated) · Recommended: 16 GB+ for comfortable context headroom (Q4-class, calculated)
How much VRAM does IQuest Q1 need at each quantization?
IQuest Q1 needs 212.9 GB of VRAM at Q4_K_M. The table below lists weights-only size and total VRAM including overhead for each common quantization level with 4k and 32k token contexts.
| Quantization | Bits / weight | Weights only | Total + KV @4k ctx | Total + KV @32k ctx |
|---|---|---|---|---|
| Q4_K_M | 4.8 | 192.2 GB | 212.9 GB | 223.2 GB |
| Q5_K_M | 5.7 | 228.2 GB | 252.5 GB | 262.9 GB |
| Q6_K | 6.6 | 264.3 GB | 292.2 GB | 302.5 GB |
| Q8_0 | 8.5 | 340.3 GB | 375.9 GB | 386.2 GB |
| FP16 | 16 | 640.6 GB | 706.2 GB | 716.5 GB |
config.json declares a sliding window of 4096 on attention layers. The card targets agentic coding/tool-use workloads; heavy long-context serving is secondary.
Method: weights = parameters × bits-per-weight (Q4_K_M ≈ 4.8, Q5_K_M ≈ 5.7, Q6_K ≈ 6.6, Q8_0 ≈ 8.5, FP16 = 16), plus 10% loading overhead, plus KV cache from the verified architecture config (88 layers, 8 KV heads, 128 head dim). Source: HF-verified 2026-10-05 via plain api (gated=false safetensors.total=320318615552 params=320.32B ctx=524288 license=other repo=IQuestLab/IQuest-Q1@5c21b0630586ef77d38cff1094b8cf37a417fcd8 layers=88 hidden=3072 kv_heads=8 heads=48 vocab=160000). MoE: num_experts=256 num_experts_per_tok=8 active_params_b=15.0 (vendor model card (Total 320B, Activated 15B per token, stated as approximate/estimated); config num_experts=256 routed 8 per token). VRAM figures are documented calculations (Q4-class ~0.6 GB per B params, on active params for MoE rows), not measurements. release_date is the HF repo lastModified date observed on the verification pass. No tok/s or benchmark scores invented.
Which GPUs can run IQuest Q1 locally?
At Q4_K_M with a 4k context, IQuest Q1 needs 212.9 GB of VRAM. The lists below are computed live from our GPU database and grouped by how much headroom the card has. Disclosure: CompareAIHardware.com participates in the Amazon Associates program and earns from qualifying purchases through links on this page. Affiliate relationships do not influence our recommendations.
Runs IQuest Q1 comfortably (25%+ VRAM headroom)
| GPU | VRAM | Bandwidth | Type | |
|---|---|---|---|---|
| AMD Instinct MI350X | 288 GB | 8000 GB/s | Data Center GPU | |
| AMD Instinct MI355X | 288 GB | 8000 GB/s | Data Center GPU | |
| NVIDIA B300 (Blackwell Ultra) | 288 GB | 8000 GB/s | Data Center GPU | |
| AMD Instinct MI430X | 432 GB | 23300 GB/s | Data Center GPU | |
| AMD Instinct MI455X | 432 GB | 23300 GB/s | Data Center GPU |
Minimum GPUs that fit IQuest Q1 at Q4_K_M
These GPUs hold the model but leave little headroom — keep contexts short.
| GPU | VRAM | Bandwidth | Type | |
|---|---|---|---|---|
| AMD Instinct MI325X | 256 GB | 6000 GB/s | Data Center GPU |
Needs 2+ GPUs to run IQuest Q1
One of these cards is too small on its own, but a pair (tensor or pipeline parallel, ~90% efficiency) covers the 212.9 GB requirement. See our multi-GPU guide for setup.
| GPU | VRAM | Bandwidth | Type | |
|---|---|---|---|---|
| H200 SXM | 141 GB ×2 | 4800 GB/s | Data Center GPU | |
| AMD Instinct MI300X | 192 GB ×2 | 5300 GB/s | Data Center GPU | |
| NVIDIA B200 | 192 GB ×2 | 8000 GB/s | Data Center GPU |
Can a Mac run IQuest Q1?
Yes — these Apple Silicon machines fit IQuest Q1 at Q4_K_M, since macOS lets the GPU use about 75% of unified memory. Generation speed is bound by memory bandwidth, so the GB/s column matters as much as capacity.
| Apple Silicon | Unified memory | Usable by GPU (~75%) | Bandwidth |
|---|---|---|---|
| M5 Ultra (Mac Studio) | 512 GB | 384 GB | 1200 GB/s |