Can a Jetson AGX Orin 64GB run Qwen3.8-Flash-Next Gyro-M?
Qwen3.8-Flash-Next Gyro-M speed by board
- Jetson AGX Orin 64GB: 25.8 tokens/s best (TQ2_0 (2-bit trellis experts), Strata Orin fork bb98d09 (CUDA SM87, MTP speculative decoding, MAXN)), 1 test
FAQ
Which edge board runs Qwen3.8-Flash-Next Gyro-M fastest?
Of the boards tested, Jetson AGX Orin 64GB is fastest for Qwen3.8-Flash-Next Gyro-M: 25.8 tokens/s generation (TQ2_0 (2-bit trellis experts), Strata Orin fork bb98d09 (CUDA SM87, MTP speculative decoding, MAXN)).
How fast does Qwen3.8-Flash-Next Gyro-M run on Jetson AGX Orin 64GB?
Best reported generation speed for Qwen3.8-Flash-Next Gyro-M on Jetson AGX Orin 64GB is 25.8 tokens/s (TQ2_0 (2-bit trellis experts), Strata Orin fork bb98d09 (CUDA SM87, MTP speculative decoding, MAXN)), from 1 sourced test. Source: kim-haneol (Strata-Orin) (https://github.com/kim-haneol/Strata-Orin/blob/main/docs/OPTIMIZATION.md).
| Board ↕ | Model ↕ | Runtime & setup ↕ | Gen t/s ↕ | Price ₹ (India) ↕ | Source ↕ |
|---|---|---|---|---|---|
| Jetson AGX Orin 64GB64 GB RAM | Qwen3.8-Flash-Next Gyro-MTQ2_0 (2-bit trellis experts) | Strata Orin fork bb98d09 (CUDA SM87, MTP speculative decoding, MAXN)NVMe | 25.8prompt 555.0 t/s | No India listing yet | Community kim-haneol (Strata-Orin)2026-10-08 |
Each row shows only what its source reported: RAM, quant, prompt speed and setup appear under the main value when published. Rows with full details are listed first, fastest first. Source badges: Community, Vendor, Reproduced; ⚠ marks a speed far from similar results (how we check).