Instinct MI100 Price & Used Prices: Which Local LLMs Can It Run?
The Instinct MI100 has 32GB of HBM2 memory with 1,229 GB/s of bandwidth. No price data yet. At Q4 quantization it runs models up to 32B parameters at roughly 31 tokens/s.
Instinct MI100 price history
Xianyu
Full Instinct MI100 specifications
What LLMs can the Instinct MI100 run?
Versions released since 2026 with up to 50B parameters| Model | 2-bit | 4-bit | 8-bit | ||||||
|---|---|---|---|---|---|---|---|---|---|
| Fits | Speed t/s | Context | Fits | Speed t/s | Context | Fits | Speed t/s | Context | |
| GemmaGemma 4 E4B | ✓ | 159.2 | 128K | ✓ | 114.2 | 128K | ✓ | 66.1 | 128K |
| QwenQwen3.5 9B | ✓ | 131.9 | 128K | ✓ | 94.6 | 128K | ✓ | 54.8 | 128K |
| GemmaGemma 4 12B | ✓ | 94 | 32K | ✓ | 69.7 | 32K | ✓ | 41.9 | 32K |
| GemmaGemma 4 26B A4B | ✓ | 125 | 32K | ✓ | 108.3 | 32K | ✓ | 80.5 | 8K |
| QwenQwen3.6 27B | ✓ | 51 | 32K | ✓ | 35.5 | 32K | ✓ | 19.9 | 8K |
| QwenQwen3.8 27B | ✓ | 51 | 32K | ✓ | 35.5 | 32K | ✓ | 19.9 | 8K |
| GLMGLM 4.7 Flash | ✓ | 161.5 | 128K | ✓ | 140.5 | 128K | Offload | 22.1 | — |
| GemmaGemma 4 31B | ✓ | 38.1 | 8K | ✓ | 27.8 | 8K | ✗ | — | — |
| QwenQwen3.6 35B A3B | ✓ | 157.1 | 128K | ✓ | 137.1 | 128K | Offload | 32.4 | — |
Fits when weights + KV cache + runtime overhead is at or below usable memory, single card. "Offload" on MoE models means it runs with expert weights in system RAM (only attention layers and KV cache stay in VRAM); that speed assumes 70 GB/s RAM bandwidth. The context column is the largest tier that fits.
Prices by platform
Buy links may carry affiliate tags; they do not change the price you pay.
FAQ
How much VRAM does the Instinct MI100 have?
The Instinct MI100 has 32GB of HBM2 memory with 1,229 GB/s of bandwidth.
Can the Instinct MI100 run a 70B model?
Not on a single card. At Q4_K_M with an 8K context a 70B dense model needs about 43GB, more than the Instinct MI100's 32GB.
How much does a Instinct MI100 cost?
No price data yet; it will appear here once the feed is live.