LLM Inference Benchmark Explorer
Qwen3.8-Flash-Next on 2× DGX Spark — NVFP4, SGLang, TP2 inference benchmark
Open this configuration in the LLM Inference Benchmark Explorer
| Model | Qwen3.8-Flash-Next |
|---|---|
| Parameters | 176B |
| Intelligence Index | 39.8 |
| Agentic Index | 53.6 |
| Device | 2× DGX Spark |
| Quantization | NVFP4 |
| Max C | 4 |
| Chat Capacity | 16 |
| Agentic Capacity | 6 |
| TP | 2 |
| DP | — |
| PP | — |
| Engine | SGLang |
| Speculative Decoding | — |
| C | TTFT (ms) | TPS (tok/s) | Status |
|---|---|---|---|
| 1 | 200 | 35.80 | PASS |
| 2 | 206 | 29.94 | PASS |
| 4 | 243 | 22.39 | PASS |
| 8 | 2671 | 13.80 | FAIL |
In OpenZeka's measurement, Qwen3.8-Flash-Next (176B parameters), served in NVFP4 format with SGLang on 2× DGX Spark (TP=2), reached a generation speed of 35.8 tok/s per request and a time to first token (TTFT) of 200 ms with a single request.
Considering the speed targets and the available KV cache capacity, the estimated capacity is 16 users for chat use and 6 for agentic use. The realistic capacity will likely fall between these two values. In scenarios dominated by coding, tool use, long workflows and multi-agent use, capacity approaches the agentic estimate; where shorter interactions, standard conversations and lighter tasks dominate, it approaches the chat estimate.
Notes: 262K context. 176B MoE, 6B active, 512 experts.
Intelligence Index and Agentic Index values are published by Artificial Analysis and are reproduced here with attribution.
In OpenZeka's measurement, Qwen3.8-Flash-Next (176B parameters), served in NVFP4 format with SGLang on 2× DGX Spark (TP=2), reached a generation speed of 35.8 tok/s per request and a time to first token (TTFT) of 200 ms with a single request.
Considering the speed targets and the available KV cache capacity, the estimated capacity is 16 users for chat use and 6 for agentic use. The realistic capacity will likely fall between these two values. In scenarios dominated by coding, tool use, long workflows and multi-agent use, capacity approaches the agentic estimate; where shorter interactions, standard conversations and lighter tasks dominate, it approaches the chat estimate.