Toast 1 released
Mixedbread · Model release
Source confirmed
Data as of September 28, 2026 · How the score is built
CurrentProprietaryReasoning
Released Aug 13, 2026131K contextMixedbread Agent model catalog
Toast 1
Decision readingToast 1 is tracked, but not publicly ranked yet. The profile exposes 0 sourced benchmark rows and leaves unsupported fields blank until a published record exists.
Released Aug 13, 2026 — see all recent releases
Toast 1 will be repriced, updated or retired. Get each notice with its source and date. Follow model changes
Share or export
Lineage
The sequence follows explicit supersedes links. Each score is estimated for that model; a relative can inform a sparse estimate but never sets a floor, so a newer release can score below an earlier one. Scores and prices remain blank when the corresponding public row or first-party rate is unavailable.
Base entry
Radar
Full release historyToast 1 release history
Radar confirmed these at the source. Use Toast 1 in your work? Explore Radar to follow supported changes and choose your alerts.
Radar
Spec sheet
Each documented value carries its source. Missing fields stay visible as not sourced or not published, rather than disappearing from the page.
- API model ID
- toast-1Mixedbread Agent model catalog
- Context window
- 131KMixedbread Agent model catalog
- Maximum output
- Not sourced yet
- Knowledge cutoff
- Not sourced yet
- Input modalities
- textMixedbread Agent model catalog
- Output modalities
- textMixedbread Agent model catalog
- Parameters
- Not disclosed by the provider
- Availability
- Mixedbread Chat Completions APIMixedbread Toast 1 announcement
- Cloud regions
- Not tracked yet
- Lifecycle
- activeMixedbread Toast 1 announcement
- API capabilities
- Tool calling, structured outputs, and batch support are not tracked yet
- Prompt caching
- Published at $0.036 per million cached input tokensMixedbread pricing
- Self-host
- Weights are not published
- Rate limits
- Not tracked yet
How to read this profile
The visual layer above carries the decisions. These notes preserve the model, ranking, coverage, and family context behind the numbers.
We track Toast 1, but no weighted text-model benchmark result is published on the site yet. This page shows the metadata and separate protocol evidence we can verify now; a BenchLM score will appear only if compatible public evaluations land.
Toast 1 is a proprietary model with a 131K context window. It uses an explicit reasoning mode, which can improve complex problem solving while adding latency and token use.
Available through Mixedbread's OpenAI-compatible Chat Completions API as `toast-1`, with text input and output, function calling, managed Store search tools, a 131K-token context window, and a 4K-token maximum output.
Official launch snapshot from Mixedbread's August 13, 2026 Toast 1 announcement and Agent model documentation. Toast 1 is a proprietary specialized search model that can run standalone or as a retrieval subagent. The published OfficeQA Pro V2 result belongs to GPT-5.6 Sol running in Codex with Toast 1, while the Harvey LAB subset compares full retrieval stacks at equal answer quality; neither is a standalone Toast 1 result. Mixedbread's standalone deep-search figure does not expose a machine-readable benchmark table with protocols that map cleanly to BenchLM keys, so this release remains unranked without copied system-level scores.
The profile has no source-displayable benchmark row yet.
Last updated September 28, 2026. Runtime fields remain blank until a sourced snapshot exists.
Questions
How does Toast 1 perform overall in AI benchmarks?
Toast 1 does not have any source-displayable benchmark rows yet, so this profile does not assign a public score or rank. Documented specifications remain visible, while score-led charts and claims stay unavailable until a published evaluation can be attached to the exact model.
Does Toast 1 have full benchmark coverage on BenchLM?
No. Toast 1 currently has 0 source-displayable rows across 486 tracked benchmark slots. The profile exposes published, non-generated evidence and leaves missing categories blank until an exact evaluation is available. Coverage describes how much was measured; it is not a penalty added to an individual benchmark result.
What is the context window size of Toast 1?
Toast 1 has a documented context window of 131K. That figure is the maximum combined prompt and retained-conversation space reported for this exact model; it is not the maximum output length. The profile keeps output limits separate because providers often publish those limits independently.