More
Submit build

Builds / 2x RTX 3090

2x RTX 3090: Inkling 2 Mini W4A16 INT4 on vLLM TP2 + Ampere overlay, 136.6 tok/s bench (Tech2wild)

Last updated:

Source: github.com ↗More source: github.com ↗More source: huggingface.co ↗📏 How measured: reported by the original source; RigBase re-checked the numbers at that source
GPU(s)
2x RTX 3090
Amazon · eBay (used) · Newegg
VRAM / unified memory (GB)
48
CPU
AMD Ryzen 9 5900X
Amazon · eBay (used) · Newegg
Motherboard
not stated
RAM
32 GB
Amazon · eBay (used) · Newegg
PSU
not stated
Other
2× RTX 3090 (Ampere) joined by an NVLink bridge (NV2); both cards on x4 PCIe links
Reported total cost
not stated
Power
240 W GPU power limit per card (stock 350 W)

Measured speed (generation tokens/sec)

ModelQuantEnginetok/sPrompt processing (tok/s)Details
Inkling 2 MiniW4A16 INT4 + DSparkvLLM TP2 + Ampere overlay, NVLink, 1 stream136.6not stated⚡ spec-decode (DSpark (num_speculative_tokens 7))context: 458,752 per request (TP2 max lane, text-only): BF16 KV, 483,850-token pool📏 bench/bench_decode.py: 6 fixed prompts, temperature 0, reasoning effort none, streaming, one request at a time; 136.6 tok/s aggregate (median 129.7) as published in Tech2wild's README. Tech2wild's calibration-free INT4 re-quantization of the routed experts (Tech2wild/Inkling-2-Mini-W4A16, 16.47 GB + 0.64 GB DSpark drafter); the stock image needs Tech2wild's Ampere overlay. The 137.5 tok/s table headline is the same lane; the 157.4 tok/s eval-turn decode (tool turns, 3.31 tokens per DSpark step) is not used. NVLink bridge on, cards capped at 240 W. 69-scenario tool eval quality 87.0/100.
Inkling 2 MiniW4A16 INT4 + DSparkvLLM TP2 + Ampere overlay, NVLink P2P off (NCCL_P2P_DISABLE=1), 1 stream100.5not stated⚡ spec-decode (DSpark (num_speculative_tokens 7))context: 458,752 per request (TP2 max lane)📏 Single-stream decode on the same lane with NCCL peer-to-peer disabled, as published in Tech2wild's README (137.5 with NVLink, +37%). Both cards report x4 PCIe links, so the README says a board with x16 slots may do better without a bridge.
Inkling 2 MiniW4A16 INT4vLLM TP2 + Ampere overlay, 1 stream89.8not stated⚡ spec-decode (off)context: not stated (TP2 quality comparison, greedy)📏 Median decode with no speculative decoding, greedy, as published in the quality table of Tech2wild's README (original NVFP4/MXFP8 checkpoint on the same cards and overlay: 87.2 tok/s). The lane table lists ~90 tok/s for the 131K vision lane, DSpark off.
Inkling 2 MiniW4A16 INT4 + DSparkvLLM TP2 + Ampere overlay, NVLink, 1 stream1192,302 @ 439,384-token prompt
est. TTFT @8k ≈ 3.6 s (calculated from measured prefill)
measured TTFT: 191 s @ 439,384-token prompt
⚡ spec-decode (DSpark (num_speculative_tokens 7))context: 439,384-token needle prompt on the TP2 max lane (458,752 per request)📏 Single-request needle test (passphrase at 50% depth in synthetic log text) as published in Tech2wild's README; decode after the prompt. Needle PASS. Other needles: ~32K at 3,068 tok/s prefill and 199,595 tokens in 59.3 s (3,367 tok/s), both PASS.

Prompt processing = prefill speed (how fast the prompt is read) as reported by the source, with the prompt length it was measured at; blank = the source does not report it. est. TTFT @8k is calculated from the measured prefill speed (8,192 ÷ prefill tok/s), shown only when prefill was measured on a prompt of 512+ tokens. It is not a measurement and ignores load time and caching. measured TTFT is the source's own time-to-first-token. Why prefill matters for coding assistants.

Numbers by Tech2wild (tonyd2wild); Thinking Machines Lab (Inkling 2 Mini, DSpark drafter); vLLM project (vllm/vllm-openai:inkling-2-mini image, Humming expert kernels, Inkling 2 Mini recipe); compressed-tensors pack-quantized layout; Tech2wild/Inkling-2-Mini-W4A16 weights on Hugging Face. RigBase read the GitHub README on 2026-11-22 and checked each figure against the text. RigBase did not re-run it. Speculative decoding (MTP/DFlash/DSpark) inflates predictable text; these rows prefer prose/code over counting peaks. Recipe license: MIT.

Source: https://github.com/tonyd2wild/Inkling-2-Mini-W4A16-RTX3090
https://github.com/tonyd2wild/Inkling-2-Mini-W4A16-RTX3090/blob/main/README.md
https://huggingface.co/Tech2wild/Inkling-2-Mini-W4A16

Blank fields = not stated in the source. We never estimate build data; GPU street prices above are separately sourced and dated.

Embed this result

Copy into a README or blog. Links back to this build. See badge docs (CC BY 4.0 attribution). The SVG does not invent speeds — numbers stay on the build page.

Markdown

[![Measured on RigBase](https://getrigbase.com/badge/tony-2x-rtx-3090-inkling-2-mini-w4a16-rtx3090.svg)](https://getrigbase.com/builds/tony-2x-rtx-3090-inkling-2-mini-w4a16-rtx3090/)

HTML

<a href="https://getrigbase.com/builds/tony-2x-rtx-3090-inkling-2-mini-w4a16-rtx3090/" rel="noopener"><img src="https://getrigbase.com/badge/tony-2x-rtx-3090-inkling-2-mini-w4a16-rtx3090.svg" alt="Measured on RigBase" height="20"/></a>

Preview: Measured on RigBase · How to use

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)

Quick answers: Can 2x RTX 3090 run Inkling 2 Mini?

Speculative-decoding rows on this page are not like-for-like with plain decode. See how much MTP and DFlash really speed things up (measured pairs).