Pinned
Introducing Toast 1, our first specialised search agent.
Toast 1 sets a new Pareto frontier for agentic search models.
Frontier search quality, across all domains, 12x faster, at 1/10th of the price.
Mixedbread
@mixedbreadai
The Retrieval Infrastructure for agents.
- New: Beating Fable 5 in deep search and helps you save tokens. Toast 1 is now live on Mixedbread API at launch pricing: Input: $0.50 → $0.30/1M Output: $1.20 → $0.72/1M Cached input: $0.06 → $0.036/1M OpenAI-compatible. New signups get $5 in credits.
- Feature: Auto file contextualization Long-doc chunks can miss context needed for retrieval. Turn on file contextualization: → every chunk embedded with global context → "ACME Q3 growth?" finds "it increased 23%" (acme.pdf) → +34% better context-aware retrieval (#1 ConTEB)
- New: improved search latency We built a query engine specifically for wholembed-v3, extended it to cover the long-query tail we see in production → query encoding p50 fell 39% → end-to-end search p50 down 21%
- New: mxbai-rerank-v3.1-listwise An upgrade to our listwise reranker. → GPT-5.6-sol (high) ranking quality at 61× the speed → up to 54% faster than v3 on long docs Built for: - recency-aware ranking - source-priority resolution - multi-step composite instructions Available
