Can NVIDIA DGX Spark (GB10) · 128 GB run Recursive Flint (40B-A2.5B)?
128 GB unified · 273 GB/s memory bandwidth · Recursive Flint (40B-A2.5B) has 39.8B parameters (2.5B active per token).
Short answer
Yes, excellent for chat and comfortable or better for every workload we test.
Best quantization for chat: MXFP4 (20 GB of weights) on Ollama. Generation: 100–140 tok/s, leaving about 99 GB free for your other apps.
How it rates for each workload
Same model and machine, different jobs. Agents and long documents send far bigger prompts than chat, so they need much more speed and memory.
| Workload | Verdict | Quant | Generation | First reply | Context | |
|---|---|---|---|---|---|---|
| Chatwith Open WebUI | Excellent | MXFP4 | 100–140 tok/s | under 1 s | 8K | Details |
| Coding questionswith Continue | Excellent | MXFP4 | 100–130 tok/s | ≈1 s | 16K | Details |
| Coding in a repositorywith Aider | Comfortable | MXFP4 | 80–110 tok/s | ≈4 s | 32K | Details |
| Agentic codingwith OpenCode | Comfortable | MXFP4 | 75–100 tok/s | ≈5 s | 32K | Details |
| Long documents (64K)with Open WebUI | Excellent | MXFP4 | 50–70 tok/s | ≈15 s | 64K | Details |
Speeds are estimates calibrated against measured benchmarks on similar hardware. “First reply” is the time to read the workload’s opening prompt. How we calculate
Other models for NVIDIA DGX Spark (GB10) · 128 GB
The best-rated models for chat on this machine.
- gpt-oss-120b≈55 tok/sExcellent
- Qwen3.6 35B-A3B60–80 tok/sExcellent
- Qwen3-Next 80B-A3B Instruct60–80 tok/sExcellent
- Gemma 4 26B-A4B45–60 tok/sExcellent
- Qwen3.5 35B-A3B60–80 tok/sExcellent
- LFM2.5 24B-A2B120–160 tok/sExcellent
Cheapest machines for Recursive Flint (40B-A2.5B)
Hardware rated comfortable or better for chat, lowest price first.
- Mac mini M4 32 GB$1,399 · 40–55 tok/sExcellent
- Desktop PC · Ryzen 9 9950X (no GPU) · 64 GB DDR5$1,400 · 20–35 tok/sExcellent
- Desktop PC · Intel Arc Pro B60 24GB · 32 GB RAM$1,500 · 95–160 tok/sExcellent
- Ryzen AI Max+ 395 mini PC / Framework Desktop · 64 GB$1,600 · 100–140 tok/sExcellent
- Desktop PC · Intel Arc Pro B60 24GB · 64 GB RAM$1,628 · 95–160 tok/sExcellent
- Mac mini M2 Pro 32 GB$1,699 · 65–90 tok/sExcellent
Same chip, other memory: 64 GB
Different tool, context size or workload?
These pages use typical settings. The full check models your exact setup.