Baseten ships DeepSeek v4.1 Flash, Reflection Beam 1.0, Nemotron Super 3.5 VL day-zero with claimed top benchmarks
baseten · x · 2026-10-21
Inference provider Baseten says three major open models launched this month went live on day zero: DeepSeek v4.1 Flash, Reflection Beam 1.0, and NVIDIA Nemotron Super 3.5 VL — each claiming first place on both Artificial Analysis benchmarks and OpenRouter production traffic. Charts show DeepSeek V4.1 Flash at 416 tok/s on Baseten (668 on FAST tier), Beam 1.0 at 532 tok/s ($1.30/$4.30 per M tokens), and Nemotron Super VL 3.5 at 545 tok/s, ahead of CoreWeave, Together AI, and Nebius. It's vendor PR, but the underlying latency, throughput, and pricing comparisons are verifiable.
Related event: Baseten day-one launch of top open models with fast inference(2 posts)→
More from Infra
- Investor: market underestimates compute demand 1,000x, heading terawatt-scale — inductionheads · 2026-10-21
- TiDB leak report: root-level org takeover bug left unfixed for 4 years, creds in Jira — arpit_bhayani · 2026-10-21
- FireTrident lets 80 coding agents share one repo, clearing Hono's 958 violations in 62s — dinasaur_404 · 2026-10-21
- xAI said to lean on ACC/AEC and copper as Chinese pluggables stay off-limits — jwt0625 · 2026-10-21
- Packed a16z SF infra event features fal, ElevenLabs, BFL and Krea engineers — venturetwins · 2026-10-21
- Modal Engineers Detail Sticky Serverless Routing for Real-Time AI Inference — blelbach · 2026-10-21