Sources · Twitter/X

@alexocheema

Active

alexocheema

Linked items 22
Errors 8
Last fetched 2026-05-14 00:42
Priority P3

Edit

Human-friendly label. Leave blank to fall back to the identifier.

Drives ingestion cadence and the default ranking weight applied to items from this source.

How often the watch loop polls this source. Minimum 5 min. Leave blank for the per-type default.

Optional. Pin a specific summarization model for items from this source.

Cancel

Recent items

The 15 most-recently-ingested items from this source, newest first.

Title Type Fetched
@alexocheema: 4 x M5 Max MacBooks clustered with RDMA: 512GB @ 2456GB/s, $20k, 560W, quiet. Find me a better deal... article 2026-05-13
@alexocheema: Perf per Watt of MacBook vs FPGA. article 2026-05-02
@alexocheema: Still very cool. Chart showing perf of various approaches on M4 Max MacBook vs FPGA. article 2026-05-02
@alexocheema: My M4 Max MacBook gets 3,756,165 tok/sec in pure C, compared to ~50,000 tok/sec with the FPGA. Try ... article 2026-05-02
@alexocheema: Londonmaxxing article 2026-04-21
@alexocheema: Running Qwen3.6 35B (vision) on 2 x M5 Max MacBook Pro with RDMA over Thunderbolt 5. It describes t... article 2026-04-21
@alexocheema: Ternus as CEO, Srouji as Chief Hardware Officer. The people who turned Apple Silicon into the leadi... article 2026-04-20
@alexocheema: Seth is using MLX to run Evo 2 on the mac. Check this out @awnihannun @angeloskath article 2026-04-18
@alexocheema: people are now sequencing their DNA at home, locally on DGX Sparks and Mac Studios. this madlad is ... article 2026-04-18
@alexocheema: Intelligence per Joule increased 18x in 16 months. This is THE metric to track for local AI progres... article 2026-04-08
@alexocheema: This is why we focus on the frontier. If closed wins, a few orgs will rent seek on intelligence. ... article 2026-04-06
@alexocheema: Yes, @arcee_ai should run well on Apple Silicon RDMA clusters with @exolabs. It’s a 398B model, 13B... article 2026-04-04
@alexocheema: Also, it’s e2e inference time with Q4_K_M and bs=1, ISL=4096, OSL=128. RTX 5090 has native support ... article 2026-04-03
@alexocheema: Direct comparison of NVIDIA RTX 5090 to M3 Ultra 👀 Small models should always be faster on the 5090... article 2026-04-03
@alexocheema: Center of the universe. article 2026-04-02