Can I run Intern S1 locally?
Intern S1 by internlm has 241B parameters (28B active per token, 128 experts, 8 used per token). At 4-bit with an 8K context it needs about 139 GB of GPU memory. It fits on 9 of 65 GPUs and Macs we track.
Memory needed by quant and context length
GB of GPU memory = weights + KV cache (FP16) + runtime overhead. Weights use real file sizes from internlm/Intern-S1-GGUF where available (marked •), otherwise parameter count × bits per weight. How we calculate.
| Quant | Quality | 4K ctx | 8K ctx | 32K ctx |
|---|---|---|---|---|
| 2-bit (Q2_K) | rough | 85.5 GB | 86.2 GB | 90.6 GB |
| 3-bit (Q3_K_M) | usable | 111 GB | 112 GB | 116 GB |
| 4-bit (Q4_K_M) | good, most popular | 138 GB | 139 GB | 144 GB |
| 5-bit (Q5_K_M) | very good | 161 GB | 162 GB | 166 GB |
| 6-bit (Q6_K) | near lossless | 185 GB | 186 GB | 190 GB |
| 8-bit (Q8_0) • | effectively lossless | 234 GB | 235 GB | 239 GB |
| 16-bit (F16/BF16) • | original weights | 439 GB | 440 GB | 445 GB |
Which GPUs and Macs run Intern S1?
Best quality quant that fits at 8K context. "Tight" = within 10% of usable memory. Speed is a ceiling from memory bandwidth (real speed is typically 50–80% of it).
Questions
How much VRAM does Intern S1 need?
About 139 GB at 4-bit (Q4_K_M) with an 8K context: 137 GB of weights (estimated), 1.5 GB of KV cache and 0.7 GB of runtime overhead. At 8-bit it needs about 235 GB.
What is the smallest GPU that runs Intern S1?
No single consumer GPU in our table fits it at Q4_K_M with 8K context; the smallest option that does is the Mac M3 Ultra 256GB (256 GB).
Can I run Intern S1 on a Mac?
Yes. At Q4_K_M with 8K context it fits on a Mac M3 Ultra 256GB or larger, using the default macOS GPU memory limit.
Model facts
| Hugging Face repo | internlm/Intern-S1 |
| Parameters | 240,709,856,128 (241B) · ~28B active (computed from config) |
| Layers / hidden size | 94 / 4096 |
| Attention | GQA; 64 heads, 4 KV heads, head dim 128 |
| KV cache per token (FP16) | 188.0 KiB |
| Max context | 65,536 tokens |
| License | apache-2.0 |
| Released on HF | Jul 24, 2025 |
| Downloads (30d) / likes | 22,240 / 257 |
Sources: Hugging Face model API · config.json (internlm/Intern-S1) · GGUF files · GPU specs from vendor pages (list). Data fetched Nov 17, 2026. Spot an error? Report it.