diff --git a/test/external/external_benchmark_all2all.py b/test/external/external_benchmark_all2all.py
new file mode 100644
index 0000000000..e278dfad73
--- /dev/null
+++ b/test/external/external_benchmark_all2all.py
@@ -0,0 +1,17 @@
+from tinygrad import Device, Tensor, TinyJit, dtypes
+from tinygrad.helpers import Timing, Context
+
+GPUS, DEPTH, SZ = 8, 4, 128 * 2**20
+WARMUP, ITERS = 3, 5
+devs = tuple(f"{Device.DEFAULT}:{i}" for i in range(GPUS))
+bufs = tuple(Tensor.empty(SZ, dtype=dtypes.uint8, device=dev).contiguous().realize() for _ in range(DEPTH) for dev in devs)
+
+@TinyJit
+def all_to_all(*srcs:Tensor): return Tensor.realize(*(src.to(dst) for i,src in enumerate(srcs) for j,dst in enumerate(devs) if i % GPUS != j))
+
+if __name__ == "__main__":
+  with Context(ALL2ALL=1, JIT_BATCH_SIZE=0):
+    for i in range(-WARMUP, ITERS):
+      with Timing("ALL2ALL ", lambda ns: f" {SZ*GPUS*(GPUS-1)*DEPTH/ns:.2f} GB/s", enabled=i>=0):
+        all_to_all(*bufs)
+        for dev in devs: Device[dev].synchronize()