Can the RTX A6000 run Ripple-65B Preview?
Yes. A new RTX A6000 ($8,363) runs Ripple-65B Preview at Q4 with room for about 116K tokens of context (224K with 8-bit KV cache), at about 17 tok/s estimate. At $492 per tok/s it is the cheapest of the 1 priced GPUs that fit it.
Yes, with room for long context. On the NVIDIA RTX A6000 (48 GB) there is room for about 116K tokens of context with the default FP16 KV cache, or 224K with an 8-bit (q8_0) KV cacheestimate: what is left after the weights and a 1 GB runtime reserve. That is 32K tokens or more, enough for long documents and code. The Q4_K_M GGUF of Ripple-65B Preview is 38.8 GB (source); with our 1.5 GB allowance for the runtime and a short context it needs about 40.3 GB.
Ripple-65B Preview fits 1 of the 18 GPUs we have a dated price for; the smallest of them has 48 GB. The RTX A6000's 768 GB/s of memory bandwidth makes it the fastest of the 1, at ~17 tok/s estimate. On the same card, the mixture-of-experts Recursive Flint 40B-A2.5B (2.5B of 39.8B active, 25.99 GB) estimates ~243 tok/s.
The cheapest way to run it per tok/s. At $8,363 (new) the RTX A6000 works out to $492 per tok/s. The cheapest complete build we priced that runs Ripple-65B Preview is our 48 GB: dual used RTX 3090 build at $4,687.95.
Last updated Oct 16, 2026. Speed is a bandwidth-based estimate (768 GB/s, NVIDIA), scaled by per-platform constants calibrated to 45 measured llama.cpp benchmarks (2025–26), not a benchmark of this pair. Max context is an estimate from the model's architecture for llama.cpp with flash attention.
Next size up: Qwen3-Coder-Next (48.53 GB at Q4_K_M) needs ~50 GB and does not fit on one NVIDIA RTX A6000. See what runs Qwen3-Coder-Next.
Every priced GPU that runs Ripple-65B Preview, by cost per token/sec
| GPU | VRAM | Speed estimate | Max context estimate | Low price | $ per tok/s |
|---|---|---|---|---|---|
| NVIDIA RTX A6000 | 48 GB | ~17 tok/s | 116K · 224K (8-bit) | $8,363 (new) | $492 |
Prices: gpuprix.com, 2026-11-02; see how prices are sourced. Prices move daily. Bar length shows value: longer is cheaper per tok/s.