llamaperf

AMD Radeon Pro W7900

AMD · 48GB · 5 reports

Is this your card?

On the AMD Radeon Pro W7900, LFM2.5 24B · 2.5B active at 8-bit, the model most run on it, generates a median of 95.0 tokens per second from one plain run (community reports, as of 22 Nov 2026).

As of 22 Nov 2026, the models most run on the AMD Radeon Pro W7900, each one size at one quant level, with the median generation speed of its plain runs (one device, one request, no speculative decoding):

Bars are medians on one scale, 0 to 95.0 tokens/s. Press ▶ to watch text arrive at a row's speed.

Engines people use on it: llama.cpp 2 (reports)

Run models on your AMD Radeon Pro W7900? Add your numbers.

Your coding agent can run a speed test on your machine and send it here with a free account. Or paste a result you've already posted.

Reported performance on the AMD Radeon Pro W7900

These reports mix GPU counts and memory setups, so a model listed here may have needed extra cards or CPU offloading. The calculator estimates which models fit in 48 GB of VRAM. Prompt processing is how fast the input is read, not how long you wait for the first word. How to measure time to first token.

Speeds are as each source reported them. GPU count, offloading and concurrent requests change them, and prompt-processing speed is input throughput, not time to first token. Check the full setup before comparing two reports.

Tone: mixedAMD hardware
generation:
13.7 tokens/s
quant:
ternary (AIM-W)

User reports BitNet b1.58 2B at 13.68 tok/s on a Radeon Pro W7900 with a self-built C inference stack, AimStack. The GPU backend loads the same 0.87GB ternary file, dequantizes weights on load and runs matmul in fp32; 256 of 256 greedy tokens matched the C engine bit-for-bit. The C engine runs the full model on a 2018 dual-core 8GB laptop at 3.20 tok/s with two workers, and its SIMD path measured 85.53 tok/s. BF16 is 2.4x faster at 32.96 tok/s but flips argmax on near-tied logits, so only FP32 passes validation. The GPU path gets the 78% memory saving but not yet a compute advantage; the UI server drives only the CPU C engine.

Nov 18, 2026
Tone: positiveAMD hardware
generation:
95.0 tokens/s
quant:
Q8 (GGUF)
agentictool-use

User reports LFM2.5-24B-A2B at ~95 t/s sustained on an AMD Radeon Pro W7900. Setup is llama.cpp with ROCm, Q8 quant, full 128k context, taking 30GB VRAM. User swapped it in for Ornith 1.5-35B in Pi Agent workflows and says it holds up well for mechanical and long-running recursive tasks; insight testing in council mode is pending.

Oct 16, 2026
AMD hardware
generation:
54.9 tokens/s
prompt processing (prefill):
982.2 tokens/s
quant:
Q8_0 (GGUF)

User asks whether Qwen3.8 27B Q8_0 numbers on a W7900 are reasonable. Table gives mean prefill 982.2 tok/s, mean generation 54.9 tok/s and mean server TTFT 2.27 s at depth 0, with 32K and 128K rows kept in features. No engine, context length or KV cache quant stated.

Oct 11, 2026
AMD hardware
generation:
48.7 tokens/s
prompt processing (prefill):
4315.7 tokens/s
quant:
F16 (GGUF)

User reports RWKV7 2.9B at 48.71 t/s generation and 4315.67 t/s prompt processing on an AMD Radeon Pro W7900. Setup is llama.cpp with F16 weights and ROCm backend, 99 GPU layers, 512-token prompt and 128-token generation. Q8_0 on ROCm reaches 58.59 t/s generation and 4033.24 t/s prompt; Vulkan backend gives 39.49 t/s (F16) and 45.21 t/s (Q8_0) generation.

Oct 8, 2026
Tone: mixedAMD hardware
generation:
63.5 tokens/s
prompt processing (prefill):
1119.0 tokens/s
quant:
UD-Q4_K_XL (GGUF)
codingcreative-writing

User reports Qwen3.8-Flash-Next at 63.5 t/s decode (code) and 1119 t/s prefill at 2.6K context on a Radeon Pro W7900 eGPU plus Strix Halo iGPU hybrid setup. Setup is llama-halo-hybrid fork with UD-Q4_K_XL quant at 128K context, MTP head enabled, dense layers on the W7900 and routed experts on the iGPU. Hybrid beats gufo iGPU-only (44.7 t/s decode code, 903 t/s prefill at 2.6K) and leaves 43-52 GiB RAM free versus 14-30. GLM-5.3-Flash UD-IQ3_XXS does 25 t/s (31.6 with MTP) and DS-V4-Flash UD-IQ3_XXS does 22.5 t/s. W7900 runs 100-104 C junction on long prefills with a locked 241 W power cap. One run each, no quality testing.

Oct 6, 2026

Get a weekly email of new AMD Radeon Pro W7900 reports and newly released models that fit it.

Email me new reports

Reported configurations

Individual observations from this page, not expected speeds or a ranking. Select a model to open its report, with the source and offloading conditions. Reported context may be a limit rather than actual input length.

Model / hardwareQuant / engineReported contextGeneration
BitNet b1.58 2B
AMD Radeon Pro W7900
ternary
AimStack
Not reported13.7 tokens/s
LFM2.5 24B (2.45B active) A2.5B
AMD Radeon Pro W7900
Q8
llama.cpp
128,00095.0 tokens/s
Qwen3.8 27B
AMD Radeon Pro W7900
Q8_0
Engine not reported
Not reported54.9 tokens/s
RWKV7 2.9B
AMD Radeon Pro W7900
F16
llama.cpp
Not reported48.7 tokens/s
Qwen3.8 125B (6B active) Flash-Next
AMD Radeon Pro W7900
UD-Q4_K_XL
llama-halo-hybrid
128,00063.5 tokens/s

Frequently asked

How fast is the AMD Radeon Pro W7900 for local LLMs?

On the AMD Radeon Pro W7900, LFM2.5 24B · 2.5B active at 8-bit, the model most run on it, generates a median of 95.0 tokens per second from one plain run (community reports, as of 22 Nov 2026). Also reported: RWKV7 2.9B at 16-bit (48.7 t/s, one plain run).