Qwen 3.8 Flash is 125B total / 6B active, 262,144 tokens on the official card. Qwen3.8 27B is 27.8B, 262,144 tokens. Q6 / Q5 / Q4 file sizes come from the listed GGUFs on each /models page, or a calculator estimate when a listing is missing. Q3 stays off. Q4 fits Qwen3.8 27B on one R9700 at 4K and spills on Qwen 3.8 Flash. Neither model was measured on the local desktop. Fit numbers are calculator estimates. llama.cpp first.
How we test
— Measured cells cite official amdai-bench/run@1 cards. Estimates stay labeled. Q3 is estimate-only. Q1/Q2 stay off.
Official card (Qwen3.8-Flash-Next): 125B total / 6B active, plus a 51B n-gram embedding llama.cpp can keep in system RAM. 48 layers, 12 sparse-attention layers (GQA 2×256) plus a 12×256 indexer cache (raw + pooled key). 36 linear layers hold a fixed state. Publisher UD-Q6_K_XL folder lists ~169 GB.
Existing GPU sheets only: R9700, RX 7900 XTX, RX 7900 XT, RX 7800 XT. llama.cpp, f16 KV.
Totals are calculator estimates. Speed is measured only on R9700 35B cells that already exist;
everything else stays not measured yet. This table lists Q6 / Q5 / Q4. Q3 stays on the calculator as a labeled estimate.
No official local desktop cell for this model × card. Not measured yet.
161.2 GB
Does not fit this GPU
not measured yet
No official local desktop cell for this model × card. Not measured yet.
167.9 GB
Does not fit this GPU
not measured yet
No official local desktop cell for this model × card. Not measured yet.
21.6 GB
Does not fit this GPU
not measured yet
No official local desktop cell for this model × card. Not measured yet.
23.4 GB
Does not fit this GPU
not measured yet
No official local desktop cell for this model × card. Not measured yet.
37.6 GB
Does not fit this GPU
not measured yet
No official local desktop cell for this model × card. Not measured yet.
Q5
UD-Q5_K_XL / UD-Q5_K_M
Estimate
150.3 GB
Does not fit this GPU
not measured yet
No official local desktop cell for this model × card. Not measured yet.
151.1 GB
Does not fit this GPU
not measured yet
No official local desktop cell for this model × card. Not measured yet.
157.8 GB
Does not fit this GPU
not measured yet
No official local desktop cell for this model × card. Not measured yet.
19.5 GB
Does not fit this GPU
not measured yet
No official local desktop cell for this model × card. Not measured yet.
21.3 GB
Does not fit this GPU
not measured yet
No official local desktop cell for this model × card. Not measured yet.
35.5 GB
Does not fit this GPU
not measured yet
No official local desktop cell for this model × card. Not measured yet.
Q4
UD-Q4_K_XL / UD-Q4_K_M
Estimate
105.6 GB
Does not fit this GPU
not measured yet
No official local desktop cell for this model × card. Not measured yet.
106.4 GB
Does not fit this GPU
not measured yet
No official local desktop cell for this model × card. Not measured yet.
113.1 GB
Does not fit this GPU
not measured yet
No official local desktop cell for this model × card. Not measured yet.
16.4 GB
Does not fit this GPU
not measured yet
No official local desktop cell for this model × card. Not measured yet.
18.1 GB
Does not fit this GPU
not measured yet
No official local desktop cell for this model × card. Not measured yet.
32.3 GB
Does not fit this GPU
not measured yet
No official local desktop cell for this model × card. Not measured yet.
System RAM
When a shopper file does not fit, the spill is system RAM.
Shopper lines stay Minimum system RAM needed and
RAM needed at full context on one R9700 at the model’s shopper quant,
32K selected and the official-card full context. Estimate unless a measured R9700 card overlays the kit.
Qwen 3.8 Flash
UD-Q6_K_XL at 4K or 32K does not fit every listed card on the calculator.
Minimum system RAM needed
192 GB kit
Estimate · 153.2 GiB raw · 32K context
RAM needed at full context
192 GB kit
Estimate · 159.9 GiB raw · 256K full context
Estimate. Offloaded weights + KV in RAM + runtime + OS/desktop + headroom, rounded up to a 192 GB DIMM kit. The local desktop ships 256 GB DDR5; measured host-RAM usage is not recorded until Local AI numbers arrive.
Breakdown
Offloaded weights ·
125.4 GiB
Separable embed in RAM ·
0 GiB
KV in RAM at selected ·
1.05 GiB
KV in RAM at full context ·
7.61 GiB
Runtime / backend ·
2.78 GiB
OS / desktop baseline ·
8.00 GiB
Headroom ·
16.0 GiB
Qwen3.8 27B
UD-Q4_K_M at 4K or 32K does not fit every listed card on the calculator.
Minimum system RAM needed
32 GB kit
Estimate · 24.7 GiB raw · 32K 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 ·
0 GiB
KV in RAM at full context ·
0 GiB
Runtime / backend ·
0.66 GiB
OS / desktop baseline ·
8.00 GiB
Headroom ·
16.0 GiB
llama.cpp first
Commands and recipes already on the model pages. Cited-model shapes are not measured local desktop recipes.
House Flash keeps its existing Strata recipe plus the llama.cpp backend page.
# Shape only — not a measured local desktop recipe.
./build-vk/bin/llama-server \
-m ~/models/Qwen3.8-27B-UD-Q4_K_M.gguf \
-ngl 99 \
-c 8192 \
--cache-type-k f16 \
--cache-type-v f16 \
--host 127.0.0.1 --port 8080
Plain-English pick
Chat
Qwen3.8 27B holds Q4 on one R9700 at 4K. Qwen 3.8 Flash needs RAM offload.
Coding agents
Keep llama.cpp first. The side that holds Q4 in VRAM (Qwen3.8 27B) is the less painful coding-agent host. House Flash / agent pins stay R9700 / local desktop — they do not transfer. No invented agent speed.
Long context
Cited context is 262,144 tokens vs 262,144 tokens. KV grows with context. Q4 at 32K spills on Qwen 3.8 Flash and fits Qwen3.8 27B on one R9700. Read /long-context for measured R9700 35B 128K–262K cards. We do not invent a 128K decode figure here.
Size and VRAM
Qwen 3.8 Flash files: UD-Q6_K_XL 169 GB, UD-Q5_K_XL 158.3 GB, UD-Q4_K_XL 111 GB. Qwen3.8 27B files: UD-Q6_K 21.98 GB, UD-Q5_K_M 19.77 GB, UD-Q4_K_M 16.46 GB. Fit on the four existing GPU sheets below. Empty speed cells stay not measured yet.
Stack Lab
Open in Stack Lab only where the handoff is already allowlisted. Doctor is always here.
House daily / Flash / agent profiles appear only when that house model is on the pair.
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
.
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
.
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
.
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.