Qwen3.7-Max input/cached-input/output prices rose from $1.50/$0.30/$4.50 to $2.50/$0.50/$7.50, increasing customer costs. NVIDIA Nemotron 3.5 ASR fell from $0.0045 to $0.0015; several model listings were removed, while Muse Spark Open and Reflection Beam labels were updated.
Machine-written from the diff below (Sol; type: price change). Check the diff before relying on it.
What changed
Struck in red is what Together AI removed; blue is what they added. Muted text is unchanged context.
Compared with the previous archived version (#15375 to #16983). Text is normalized before comparing: dates, counters and tokens show as ‹date›, ‹count›, ‹tok›, and noise-only edits are hidden. Rewindex shows changed passages with a few words of context, not a copy of the page. View the live page (opens in new tab).
Archive proof
Rewindex commits each page capture it captures to a daily checkpoint, chains it to the day before, and has the checkpoint timestamped by FreeTSA and DigiCert. That proves these records existed by then and have not been changed since, by anyone, Rewindex included.
Qwen3.7-Max input/cached-input/output prices rose from $1.50/$0.30/$4.50 to $2.50/$0.50/$7.50, increasing customer costs. NVIDIA Nemotron 3.5 ASR fell from $0.0045 to $0.0015; several model listings were removed, while Muse Spark Open and Reflection Beam labels were updated.
Together AI added MiniMax M3.1 Flash pricing at $0.15, $0.015 (cached), and $0.60, and removed pricing listings for Gemma 4 31B, DeepSeek V4 Flash 0731, Llama 3.3 70B, and MiniMax M2.7. Customers comparing models have a new priced option; the removed listings do not establish whether those models remain available.