Today's price · per 1M tokens
Input
$0.950
per 1M tokens
Output
$4.00
per 1M tokens
Blended
$2.48
avg of input & output $/1M
Cached input
$0.190
20% of input — prompt caching
Overview
Moonshot's coding-focused 1T-parameter MoE model with a 256K context and forced 'thinking' mode, tuned for agentic coding with ~30% lower reasoning-token use than K2.6. OpenAI-compatible API, Modified-MIT weights. $0.95/$4.00 per 1M tokens.
Deploy this open model on rented GPUs
Open weights mean you can self-host instead of paying per-token API prices. These platforms let you serve it on demand — often cheaper at scale.
Partner links — we may earn a commission if you sign up, at no cost to you. We only list platforms we'd recommend regardless.
Capabilities
struck through = not supportedBenchmark performance
accuracy % · higher is betterEvery per-benchmark score we hold for this model is reported by its own vendor, so none is counted toward an accuracy average or any ranking.
80 tokens/sec output
Source: Moonshot model card · updated May 1, 2026 · See full rankings →
Specifications
- Provider
- Moonshot
- Context window
- 262K tokens
- Modality
- text, image, video
- Parameters
- Proprietary
- Open source
- Yes — open weights available
- Released
- Jun 12, 2026
- Status
- Current
- Last updated
- Jun 25, 2026
- Tags
Availability verified: Not re-verified. We have no recent evidence that Moonshot still lists this model — treat availability as unconfirmed.