RigPerDollar

Can the Radeon RX 7900 XTX run Meta Muse Glimmer 30B?

Yes. A new Radeon RX 7900 XTX ($1,350) runs Meta Muse Glimmer 30B at Q4 with room for its full 128K context, at about 40 tok/s estimate. At $34 per tok/s it is the cheapest of the 5 priced GPUs that fit it.

Yes, with room for long contextMeta Muse Glimmer 30B Q4_K_M on the AMD Radeon RX 7900 XTX
VRAM needed~19.9 GB18.39 GB file + allowance
Max context estimate128K tokens (full)FP16 KV cache
Lowest price$1,350 newsource, 2026-11-02
Cost per tok/s estimate$34#1 of 5 priced GPUs

Yes, with room for long context. On the AMD Radeon RX 7900 XTX (24 GB) there is room for about its full 128K tokens of context with the default FP16 KV cacheestimate: what is left after the weights and a 1 GB runtime reserve. The Q4_K_M GGUF of Meta Muse Glimmer 30B is 18.39 GB (source); with our 1.5 GB allowance for the runtime and a short context it needs about 19.9 GB.

Meta Muse Glimmer 30B fits 5 of the 18 GPUs we have a dated price for; the smallest of them has 24 GB. The Radeon RX 7900 XTX's 960 GB/s gives ~40 tok/s estimate, 51% of the fastest card, the GeForce RTX 5090 at ~78 tok/s. On the same card, the mixture-of-experts K2-Horizon-MoVA-36B-A4B (4B of 37.44B active, 22.37 GB) estimates ~178 tok/s.

The cheapest way to run it per tok/s. At $1,350 (new) the Radeon RX 7900 XTX works out to $34 per tok/s; the next best is the GeForce RTX 3090 at $35. The cheapest complete build we priced that runs Meta Muse Glimmer 30B is our 24 GB: one used RTX 3090 build at $2,650.58.

Buy on AmazonAs an Amazon Associate I earn from qualifying purchases.

Last updated Oct 16, 2026. Speed is a bandwidth-based estimate (960 GB/s, AMD), 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: NVIDIA Nemotron 3.5 Lightning 30B-A3B (25.48 GB at Q4_K_M) needs ~27 GB and does not fit on one AMD Radeon RX 7900 XTX. See what runs NVIDIA Nemotron 3.5 Lightning 30B-A3B.

Every priced GPU that runs Meta Muse Glimmer 30B, by cost per token/sec

GPUVRAMSpeed estimateMax context estimateLow price$ per tok/s
AMD Radeon RX 7900 XTX24 GB~40 tok/s128K (full)$1,350 (new)$34
NVIDIA GeForce RTX 309024 GB~43 tok/s128K (full)$1,499 (used)$35
NVIDIA GeForce RTX 409024 GB~46 tok/s128K (full)$2,910 (used)$63
NVIDIA GeForce RTX 509032 GB~78 tok/s128K (full)$6,500 (new)$83
NVIDIA RTX A600048 GB~35 tok/s128K (full)$8,363 (new)$239

Prices: picksbycard.com, 2026-11-02; see how prices are sourced. Prices move daily. Bar length shows value: longer is cheaper per tok/s.

Related