More
Submit build

Models / Ripple 65B Preview

Best home rig for Ripple 65B Preview

Quick answer

How much VRAM does Ripple 65B Preview need?
The weights are about 39.8 GB at Q4_K_M (65B params × 4.89 bits ÷ 8, before context/KV cache). Add memory for context (KV cache) and the runtime. The smallest-memory setup with a measured result here, fully in memory, is Apple M3 Ultra (96 GB) at 15.0 tok/s.

What is the fastest measured hardware for Ripple 65B Preview?
RTX 3090: 25.9 tok/s generation (llama.cpp b12167 llama-server (CUDA), AD-IQ2_XXS (AtomicChat GGUF, 20.0 GB)), from https://inferencekey.com/research/ripple-65b-dense-coder-one-card.

Last updated:

Current model65B dense4 measured resultsfastest plain decode: 25.9 tok/s model info ↗

Will it fit my GPU? Best value GPU per euro →

Real, measured generation speeds for Ripple 65B Preview on home hardware, fastest first. Different quants/engines aren't directly comparable.

BuildQuantEnginetok/sPrompt processing (tok/s)CostSource
RTX 3090 24 GB: Ripple-65B Preview 2-bit AD-IQ2_XXS on llama.cpp at 23.6-30.9 tok/s (16K ctx, one wrong test in three coding answers)AD-IQ2_XXS (AtomicChat GGUF, 20.0 GB)llama.cpp b12167 llama-server (CUDA)25.9–inferencekey.com ↗
Mac Studio M3 Ultra 96 GB: Ripple-65B Preview Q4_K_M on llama.cpp at 15.0 tok/s (13.5 after a 32K prompt, 9.45 at 128K)Q4_K_M (self-quantized, 38.8 GB)llama.cpp (Metal)15.0126 @512 · est. TTFT @8k ≈65.1 s*leanzero.net ↗
RTX A6000 48 GB: Ripple-65B Preview (dense 65B coder) Q4_K_M on llama.cpp at 14.3-16.3 tok/s, prompts up to 256K on one cardQ4_K_M (AtomicChat GGUF, 38.8 GB)llama.cpp b12167 llama-server (CUDA)14.4540 @25,930 · est. TTFT @8k ≈15.2 s*inferencekey.com ↗
4x Intel Arc Pro B70 (Xeon workstation): Ripple-65B Preview Q4_K_M on llama.cpp SYCL at 13.0 tok/s, layer split, 200K contextQ4_K_M (AtomicChat GGUF, 38.8 GB)llama.cpp b12314 (SYCL), layer split across 4 GPUs13.0–intelinside.ai ↗

Prompt processing = prefill speed reported by the source, @ prompt length. * est. TTFT @8k is calculated from measured prefill, not measured. Why it matters.

Quality cost of each quant for Ripple 65B Preview

How far each quantization drifts from the full-precision (BF16) model, as published by the quant maker. Lower KL divergence (KLD) is closer to the original; same-top-token % is how often the quant picks the same next token as BF16. Q4_K_M is usually the sweet spot; below Q4 the drift grows fast, which is why RigBase only lists Q4-or-higher speed results.

QuantFile sizeMean KLD vs BF16Same top token
Q8_068.8 GB0.00199.0%
AD-Q6_K57.7 GB0.001698.7%
AD-Q5_K_M47.9 GB0.006797.3%
AD-Q4_K_M40.4 GB0.013695.9%
AD-IQ4_XS38.7 GB0.014995.6%
Q4_K_M (plain llama.cpp, no imatrix)38.8 GB0.03893.0%
AD-IQ3_S below Q432.4 GB0.032793.4%
AD-IQ3_XXS below Q428.0 GB0.058491.2%
AD-IQ2_S below Q425.3 GB0.095988.5%
AD-IQ2_XS below Q422.1 GB0.177884.2%
AD-IQ2_XXS below Q420.0 GB0.260580.8%
AD-IQ1_M below Q418.5 GB0.374376.8%

Source: huggingface.co ↗. AtomicChat (quant publisher) GGUF card: llama-perplexity on their 30-language neutral corpus (87 chunks of 4,096 tokens), llama.cpp 9e2a04a with the Ripple port, vs the BF16 original. AD files are AtomicChat's imatrix per-tensor layouts; the plain Q4_K_M is a no-imatrix baseline. KLD from different publishers, datasets or llama.cpp builds isn't directly comparable.

Rent before you buy: try this model on a cloud GPU for a few dollars first: Runpod (referral link; sign up with Google to get the new-user bonus)

Memory math: how much VRAM / RAM does Ripple 65B Preview need?

Ripple 65B Preview has 65B parameters (source: Ricursive model card: dense 65B, 80 layers (64 sliding-window + 16 global), Apache 2.0, released 4 Nov 2026). Weight size = parameters × bits-per-weight ÷ 8, before context (KV cache) and runtime overhead.

FormatBits/weightWeightsSmallest common memory size that holds the weights alone
Q8_08.5069.1 GB96 GB
Q6_K6.5653.3 GB64 GB
Q5_K_M5.7046.3 GB48 GB
Q4_K_M4.8939.8 GB48 GB
Q3_K_M4.0032.5 GB48 GB
Q2_K3.1625.7 GB32 GB

Bits-per-weight for GGUF quants are llama.cpp's own figures (llama.cpp quantize README). "As measured" rows apply the source's stated bits-per-weight to the parameter count; real files differ slightly because some layers stay at higher precision. Try other combinations in the VRAM calculator.

Measured on setups with less GPU memory than the Q4_K_M weights (so part of the model lived in system RAM or was streamed from disk):

FAQ

How much VRAM does Ripple 65B Preview need?

The weights are about 39.8 GB at Q4_K_M (65B params × 4.89 bits ÷ 8, before context/KV cache). Add memory for context (KV cache) and the runtime. The smallest-memory setup with a measured result here, fully in memory, is Apple M3 Ultra (96 GB) at 15.0 tok/s.

What is the fastest measured hardware for Ripple 65B Preview?

RTX 3090: 25.9 tok/s generation (llama.cpp b12167 llama-server (CUDA), AD-IQ2_XXS (AtomicChat GGUF, 20.0 GB)), from https://inferencekey.com/research/ripple-65b-dense-coder-one-card.

Which GPUs have measured results for Ripple 65B Preview?

4x Intel Arc Pro B70, Apple M3 Ultra, RTX 3090.

Can it run on… 4x Intel Arc Pro B70 · Apple M3 Ultra · RTX 3090

Related: what GPU do you need for 2026 models · all builds

Search answer: How much VRAM do you need for Ripple 65B Preview? · Will it fit my GPU?