Fable-tier, 100 tok/s.
Opus-tier. 744B MoE, 1M context.
V4.1 Flash. Vision, 1M context.
DeepSeek now, Kimi soon. Approval required.
Typed decisions from text, JSON or images. Probabilities, no output tokens.
Merge AI-generated code edits instantly.
AI search subagent with sub-6s searches.
Verbatim context compaction for long-running agents.
Auto-route each prompt to the best model.
Unlimited Kimi K3.1 and DeepSeek. $200 a month.
Reserved B200 capacity by the GPU-hour.
Per-token model pricing.
GB300, B300, B200 by the GPU-hour.
Engineering deep dives and product updates.
Up to $1K in API credits for startups.
Talk to the team about your use case.
How we train and deploy models.
We are not hiring.
WarpGrep v2 is an RL-trained parallel search subagent that lifts every major coding model to #1 on SWE-Bench Pro. 15.6% cheaper, 28% faster, and now handling multi-repo, package, and log search.