Unverified50% confidenceBenchmarkExact time
An author chained together two gaming PCs, an M4 Pro MacBook and two iPhone 17 Pro Max devices to run Kun 3.8 Flash Next (a 48-layer MoE model with 512 experts per layer), measuring 37-39 tokens/s at a 60k context.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
10/18/2026
First Seen
Valid until: 11/1/2026
Sources
Related Entities
Related Claims
UnverifiedA developer connected an iPhone 17 Pro Max to a 24GB-memory M4 Pro MacBook via a 10 Gb/s USB-C cable, using the phone as a second GPU to share the computational load for a locally running Qwen 3.8 27B (IQ4_XS quantization) model.73% similarUnverifiedLocal inference is constrained by the user's hardware: a 7B-parameter quantized model (such as the Q4 version of Llama 3 8B) has a file size of approximately 4–5 GB, can reach 20–40 tokens per second on a Mac with an Apple M-series chip, and may drop to 3–5 tokens per second on a laptop with only an integrated GPU and 8 GB of RAM.70% similarUnverifiedThe Qwen3.8 Flash Next (Q3 quantized version) weights are approximately 84 GB and run on an AMD 7900XTX 24 GB GPU with an average decoding output speed of approximately 90 tokens/s.68% similarUnverifiedOn a Mac Studio (Apple M3 Ultra, 96 GB memory), in tests using mflux 0.19.1, 8-bit quantized weights, and 1600x896 resolution, ERNIE-Image Turbo averaged a score of 4.28 (out of 5) while FLUX.2 klein 4B averaged 3.07.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1059795API
curl https://kongchang.com/api/v1/knowledge/claims/1059795MCP
get_claim(id=1059795)