Upstream UpdateNullstack contributes Spectrum diffusion & CDNA3 attention upstream to SGLangRead PR log
Nullstackby NullStack
Pricing

Choose a plan.

Pick a plan for private coding-agent inference, then scale with usage packs or a team plan as usage grows.

Most Popular

Pro

For individual developers & engineers

$20/month

~300M tokens expected monthly usage

Included Features
  • GLM-5.2, DeepSeek V4 Flash 0731, and Kimi K3
  • OpenAI-compatible REST API (/v1)
  • Nullstack CLI launcher with automated privacy
  • Web search at $0.02 per search
  • Usage packs always available
  • Zero request retention guarantee
  • No training on your requests or code
5x Usage

Max

For high-frequency multi-agent workflows

$60/month

~1.5B tokens expected monthly usage

Included Features
  • Everything in Pro
  • 5x more usage for heavy multi-turn coding sessions
  • Multiple concurrent agent threads
  • Priority GPU queueing during peak hours
  • Usage packs always available
  • Zero request retention guarantee
  • No training on your requests or code
Custom SLA

Enterprise

For teams, orgs, and compliance requirements

Custom

Custom pooled usage plan

Included Features
  • Team and organization multi-seat workflows
  • Shared API-key planning and spend limits
  • Custom usage plan and dedicated GPU allocation
  • Guaranteed custom SLA & 99.99% uptime
  • Direct Slack / Discord support channel
  • Enterprise security & procurement review (SOC 2, ISO 27001)
  • On-prem and private VPC deployment options

Web search tool

Agent web search is billed per search query. Available across all plans. Prompts and completions are never sent to Brave Search—only the standalone search query is transferred.

$0.02per web search

Token Pricing by Model

Direct pay-as-you-go rates across input, output, and cache read tokens.

ModelContextInput / 1MOutput / 1MCache Read / 1M
GLM-5.2(Z.ai)
524K$1.10$4.00$0.20
DeepSeek V4 Flash 0731(DeepSeek)
1M$0.14$0.28$0.0028
Kimi K3(Moonshot AI)
1M$2.50$12.00$0.45
Laguna 2.1(NullStack)
2MComing soon

Usage Packs

Add usage at any time. Usage packs never change your plan terms, and remain valid for 90 days.

$5

Instant top-up for light overflow testing.

90-day expiry
$10

Standard extension for agent refactors.

90-day expiry
$20

Popular pack for extensive multi-day runs.

90-day expiry
$50

Heavy workload booster for repo migrations.

90-day expiry
$100

Maximum pack for autonomous agent fleets.

90-day expiry

Interactive Spend Estimator

Estimate monthly costs with prefix caching discounts for coding sessions.

Presets:
Monthly Token Volume150 Million tokens
10M tokens1,000M (1 Billion)2,000M tokens
Prefix Caching Rate (Multi-turn Sessions)40% cached

Coding agents typically achieve 40% - 75% prefix cache hits on large multi-file prompts.

Estimated Monthly Cost
$257.70/ month
Prefix Caching saves ~$37.80/mo
Uncached Input$69.30
Cached Input (Discounted)$8.40
Output Tokens$180.00