Which quant should I download?

Prefer the shopper Q4_K_M file when it still fits. Q6 / Q5 / Q4 are the listed shopper steps. Q3_K_M is a labeled estimate — not a house pick. Q1/Q2 stay off. See Quants, Q6 / Q5 / Q4 + Q3 estimate.

Measured decode is only the official local desktop 35B llama.cpp Vulkan sweep on one or two R9700s. Every other model × card stays not measured yet. Record a card with amdai-bench. See Contribute, amdai-bench, Benchmarks.

Can my Radeon run it? · RX 6700 XT · gpt-oss 20B · llama.cpp

Can RX 6700 XT run gpt-oss 20B?

Needs 2 GPUs

RX 6700 XT (12 GB) × gpt-oss 20B. Shopper file is Q4_K_M at 256K. Calculator verdict: needs 2 gpus. llama.cpp first (Vulkan, then HIP). Q3 is a labeled estimate, not a house pick. Measured decode appears only where a local desktop card already exists. We do not invent a price.

Next step

Your next step for this model: One Radeon AI PRO R9700 (32 GB) fits this model fully on one card.

Or add a second AMD Radeon RX 6700 XT (see the 2x columns)

Newegg's R9700 offers have been marketplace sellers (checked 2026-10-30), so that seller's return policy applies, not Newegg's. Open the listing's Warranty & Returns tab before you buy.

A commission does not change the fit verdict on this page. We do not print a price.

As an Amazon Associate I earn from qualifying purchases. Some Newegg links use Rakuten Advertising and some B&H links use B&H’s affiliate program when those IDs are configured. Micro Center and AMD Store links are plain “also available at” links — no tracking. Buttons say “Check price” and never invent or live-quote a price. Full disclosure.

How we test — Measured cells cite official amdai-bench/run@1 cards. Estimates stay labeled. Q3 is estimate-only. Q1/Q2 stay off.

Q6 / Q5 / Q4 plus a Q3 estimate

Weights are official GGUF listings when the model page has them. Totals include KV at f16. 1-GPU and 2-GPU columns use the same calculator as this share URL.

gpt-oss 20B on RX 6700 XT fit table
Quant	Weights	1× 4K	1× 32K	1× 128K	2× 4K	2× 32K	2× 128K	Speed
Q6_K Estimate	11.2 GB Hugging Face LFS size for unsloth/gpt-oss-20b-GGUF gpt-oss-20b-Q6_K.gguf (12.04 GB)	Tight — little desktop headroom 11.9 GB	Does not fit this GPU 12.6 GB	Does not fit this GPU 14.8 GB	Fits in VRAM (estimate) 5.95 GB/GPU	Fits in VRAM (estimate) 6.28 GB/GPU	Fits in VRAM (estimate) 7.42 GB/GPU	not measured yet No official local desktop cell for this model × card. Empty stays not measured yet. Contribute · amdai-bench
Q5_K_M Estimate	10.9 GB Hugging Face LFS size for unsloth/gpt-oss-20b-GGUF gpt-oss-20b-Q5_K_M.gguf (11.72 GB)	Tight — little desktop headroom 11.6 GB	Does not fit this GPU 12.3 GB	Does not fit this GPU 14.5 GB	Fits in VRAM (estimate) 5.80 GB/GPU	Fits in VRAM (estimate) 6.13 GB/GPU	Fits in VRAM (estimate) 7.27 GB/GPU	not measured yet No official local desktop cell for this model × card. Empty stays not measured yet. Contribute · amdai-bench
Q4_K_M Estimate	10.8 GB Hugging Face LFS size for unsloth/gpt-oss-20b-GGUF gpt-oss-20b-Q4_K_M.gguf (11.62 GB)	Tight — little desktop headroom 11.5 GB	Does not fit this GPU 12.2 GB	Does not fit this GPU 14.4 GB	Fits in VRAM (estimate) 5.75 GB/GPU	Fits in VRAM (estimate) 6.08 GB/GPU	Fits in VRAM (estimate) 7.22 GB/GPU	not measured yet No official local desktop cell for this model × card. Empty stays not measured yet. Contribute · amdai-bench
Q3_K_M Q3 estimate	10.7 GB Hugging Face LFS size for unsloth/gpt-oss-20b-GGUF gpt-oss-20b-Q3_K_M.gguf (11.51 GB). Q3 is a labeled estimate — not a local desktop run	Fits in VRAM (estimate) 11.4 GB	Does not fit this GPU 12.1 GB	Does not fit this GPU 14.3 GB	Fits in VRAM (estimate) 5.70 GB/GPU	Fits in VRAM (estimate) 6.03 GB/GPU	Fits in VRAM (estimate) 7.17 GB/GPU	not measured yet Q3_K_M is a labeled estimate — not a local desktop run. Contribute · amdai-bench

System RAM

When the file does not stay in 12 GB VRAM, the spill is system RAM. Shopper lines stay Minimum system RAM needed and RAM needed at full context for Q4_K_M on this card. Estimate unless a measured R9700 card overlays the kit.

Minimum system RAM needed

32 GB kit

Estimate · 29.5 GiB raw · 256K context

RAM needed at full context

32 GB kit

Estimate · 24.6 GiB raw · 8K full context

Estimate. Offloaded weights + KV in RAM + runtime + OS/desktop + headroom, rounded up to a 32 GB DIMM kit.

Breakdown

Offloaded weights ·: 0 GiB
Separable embed in RAM ·: 0 GiB
KV in RAM at selected ·: 4.84 GiB
KV in RAM at full context ·: 0 GiB
Runtime / backend ·: 0.65 GiB
OS / desktop baseline ·: 8.00 GiB
Headroom ·: 16.0 GiB

llama.cpp launch

Vulkan first, then ROCm/HIP. Copy a house recipe when one exists. llama.cpp stays first. Strata / vLLM stay on /backends.

llama.cpp Vulkan

gpt-oss 20B llama.cpp

Shape only — not a measured local desktop recipe. Copy the cited GGUF name from the model page.

# llama.cpp Vulkan. House pin when a recipe exists; otherwise a shape only.
./build-vk/bin/llama-server \
  -m ~/models/gpt-oss-20b-Q4_K_M.gguf \
  -ngl 99 \
  -c 262144 \
  --cache-type-k f16 \
  --cache-type-v f16 \
  --host 127.0.0.1 --port 8080

llama.cpp HIP

ROCm / HIP after rocminfo

HIP decode is not measured yet. Confirm gfx with rocminfo first.

# HIP shape only — house 35B cards are Vulkan. HIP decode is not measured yet.
./build-hip/bin/llama-server \
  -m ~/models/gpt-oss-20b-Q4_K_M.gguf \
  -ngl 99 \
  -c 262144 \
  --cache-type-k f16 \
  --cache-type-v f16 \
  --host 127.0.0.1 --port 8080

Same model, other cards

gpt-oss 20B on the other published Radeon sheets.

Other models on RX 6700 XT

Check price

Existing /go/ routes for this GPU sheet. Buttons say Check price — we do not print a figure.

Check price · no live or invented prices

Radeon RX 6700 XT

12 GB RDNA 2 XT listing. Not the 16 GB RX 6800 XT and not the 10 GB non-XT 6700. No live or invented price.

Return terms vary by seller and item. Newegg's Hassle-Free page covers only Sold-and-Shipped-by-Newegg listings; marketplace items use the seller's policy. See the listing's Warranty & Returns tab.

FAQ

Can RX 6700 XT run gpt-oss 20B?: Needs 2 GPUs on the calculator at the shopper Q4_K_M file and 256K. Q6 / Q5 / Q4 are listed. Q3_K_M is a labeled estimate. llama.cpp first.; This pair · RX 6700 XT · gpt-oss 20B · Calculator
Which quant should I download?: Prefer the shopper Q4_K_M file when it still fits. Q6 / Q5 / Q4 are the listed shopper steps. Q3_K_M is a labeled estimate — not a house pick. Q1/Q2 stay off.; Quants · Q6 / Q5 / Q4 + Q3 estimate
How fast is it?: Measured decode is only the official local desktop 35B llama.cpp Vulkan sweep on one or two R9700s. Every other model × card stays not measured yet. Record a card with amdai-bench.; Contribute · amdai-bench · Benchmarks

Stack Lab

Measure the pair on your box. Doctor and planner are allowlisted. House daily / Flash profiles appear only for those house models.

Open in Stack Lab

If the app is not installed, packages are not published yet. There is no public download URL. Read the Stack Lab install notes . If Stack Lab is already running on this machine, open the local app .

Open in Stack Lab

If the app is not installed, packages are not published yet. There is no public download URL. Read the Stack Lab install notes . If Stack Lab is already running on this machine, open the local app .

Also: All pairs · RX 6700 XT · gpt-oss 20B · Official GGUF · Official card · llama.cpp · Speed up llama.cpp · Dual / multi-GPU

Next steps

Real pages on this site. We do not invent a next click.

AMDai Lab Notes digest

Email only — no account. Double opt-in. About once a week, and only when run cards, guides, ROCm problems, or Stack Lab actually changed. Not a daily drip. Issues may include one labeled Sponsored mention. Slots and rates: Sponsor.

Topics New measured run cards Official amdai-bench/run@1 cards as they land on /results. Guides and tutorials New or updated getting-started notes, tutorials, and blog posts. ROCm problem updates Cited /rocm-problems rows when status or workaround changes. Stack Lab releases Radeon Stack Lab companion notes — Ubuntu 24.04+, packages not published yet.