No credit card required
Start free,
scale as you grow
Usage-based pricing across 6 dimensions. Generous free tiers on every product. Most teams pay nothing.
Starting at
$0 /month
Free tier - resets monthly
Tracing
50 GB
Evaluations
2K credits
Command Center
100K req
Guardrails
15 built-in
Simulation
1M tokens
Falcon AI
2K credits
Datasets
Unlimited
Prompts
Unlimited
Annotations
3 queues
Dashboards
Unlimited
Alerts
3 monitors
traceAI
30+ libs
Included on every plan
Unlimited team members
Unlimited projects
Full API + SDKs
BYOK evals ($0 cost)
11 span types, 70+ filters
Agent graph & sessions
Prompt versioning & labels
Agent playground
5 annotation label types
15 built-in guardrails
Datasets persist forever
MCP server
Free
No credit card required
Generous monthly free tier
Community support
Unlimited projects
Unlimited team members
30-day data retention
Storage 50 GB GB
AI Credits 2K credits
Gateway 100K requests
Cache Hits 100K hits
Text Simulation 1M tokens
Voice Simulation 60 min minutes
Region
2,400+ teams use Future AGI
Pricing
Usage-based, transparent
Every feature starts free. Only pay when you exceed the generous free tier. Set billing limits so you never get a surprise bill.
Tracing & Observability End-to-end traces, spans, sessions, dashboards, alerts 50 GB/mo
Tracing & Observability End-to-end traces, spans, sessions, dashboards, alerts
50 GB/mo
Included free
50 GB storage - traces, spans, eval results, sessions
11 span types · 70+ filters · full-text search
Sessions, end users, agent graph visualization
Custom dashboards - unlimited, drag-and-drop
3 alert monitors with Slack + email
30-day data retention
Pricing
First 50 GB Free
50 – 500 GB $2/GB
500 GB – 2 TB $1.50/GB
2 TB+ $1/GB
Includes traces, spans, eval results, dataset & annotation storage.
Evaluations Heuristic, code evals, LLM-as-judge, agentic evaluation 2K credits/mo
Evaluations Heuristic, code evals, LLM-as-judge, agentic evaluation
2K credits/mo
Included free
Heuristic evals - always free, unlimited (BLEU, ROUGE, regex)
BYOK LLM-as-judge - your own key, $0 platform cost
Code evals - charged per run via AI Credits
LLM-as-judge (Turing) - charged per run via AI Credits
7 agentic eval agents (error localizer, RAG, tool, prompt)
Eval templates, CI/CD, inline, continuous + historical
Pricing
Heuristic evals Always free
First 2K credits Free
Code & LLM-as-judge Per run (AI Credits)
2K+ credits $10/1K
Heuristic evals are always free. Code evals and LLM-as-judge (Turing models) are charged per run via AI Credits. BYOK = $0.
Command Center AI gateway - routing, caching, guardrails, cost tracking 100K req/mo
Command Center AI gateway - routing, caching, guardrails, cost tracking
100K req/mo
Included free
100K requests + 100K cache hits/mo
15+ LLM providers - OpenAI, Anthropic, Gemini, Bedrock
9 load balancing strategies, auto-failover, retry
Exact + semantic caching (6 backends)
15 built-in guardrails - always $0
Rate limiting, budgets, virtual keys, cost tracking
Pricing
First 100K requests Free
100K – 1M $5/100K
1M – 10M $4/100K
10M+ $2.50/100K
Cache hits priced separately at ~80% less. Volume discounts at scale.
Guardrails (Protect) ML content moderation, PII, injection, toxicity detection 15 built-in, $0
Guardrails (Protect) ML content moderation, PII, injection, toxicity detection
15 built-in, $0
Included free
15 rule-based guardrails - always $0 (PII, secrets, regex)
ML Protect (Gemma 3n) - uses AI Credits per check
Protect Flash: ~1-3 credits · Full: ~3-8 credits
15 external integrations (Lakera, Presidio, Llama Guard)
Enforce / Monitor / Log modes, score thresholds
PII remediation, custom blocklists
Pricing
Rule-based (15 types) Always free
Protect Flash ~1-3 credits
Protect Full ~3-8 credits
External (BYOK) $0 platform
ML Protect uses the shared AI Credits pool ($10/1K, 2K free/mo).
Simulation Text + voice agent testing with personas and scenarios 1M tokens + 60 min
Simulation Text + voice agent testing with personas and scenarios
1M tokens + 60 min
Included free
Text simulation - 1M tokens/mo (~200 short runs)
Voice simulation - 60 minutes/mo
28+ pre-built personas + custom personas
3 scenario types (script, dataset, graph)
Transcripts, session comparison, tracking
AI scenario generation (AI Credits)
Pricing
First 1M tokens Free
1M – 10M tokens $2/1M
10M+ $1.50/1M
Voice: first 60 min Free
Voice: 60+ $0.08/min
Text and voice billed separately.
Datasets & Experiments Dataset management, versioning, experiments, synthetic data Unlimited
Datasets & Experiments Dataset management, versioning, experiments, synthetic data
Unlimited
Included free
Unlimited datasets - stored forever, 14 column types
6 creation methods (manual, CSV, PDF, HuggingFace, SDK)
Unlimited experiments - multi-model comparison
Import/export CSV, Excel, PDF, DOCX
BYOK model calls - $0 platform cost
Synthetic data generation (AI Credits)
Pricing
Datasets Always free
Storage Toward Storage ($2/GB)
Synthetic data AI Credits ($10/1K)
Datasets persist forever on all plans.
Prompt & Agent Workbench Prompt management, versioning, agent builder, playground Unlimited
Prompt & Agent Workbench Prompt management, versioning, agent builder, playground
Unlimited
Included free
Prompt templates - versioned, labels, folders
Agent playground - visual graph builder, 3 node types
Prompt playground - BYOK, $0 platform cost
100+ models, full parameter control
Tool/function calling, format validation
AI-powered creation + improvement (AI Credits)
Pricing
All features Always free
BYOK model calls $0 platform
AI-powered features AI Credits
Annotations & Scoring Annotation queues, labels, review workflow 3 queues
Annotations & Scoring Annotation queues, labels, review workflow
3 queues
Included free
3 queues (Free), 10 (Boost), unlimited (Scale)
5 label types - numeric, text, categorical, star, thumbs
Unlimited annotators - $0, no per-seat cost
Auto-assign, round-robin, load-balanced
Multi-source items (traces, spans, datasets)
Review workflow (Boost+)
Pricing
Annotations Always free
Free/PAYG 3 queues
Boost ($250/mo) 10 queues
Scale ($750/mo) Unlimited
Annotations stored forever on all plans.
Falcon AI AI-powered error analysis, insights, auto-tagging, reports 2K credits/mo
Falcon AI AI-powered error analysis, insights, auto-tagging, reports
2K credits/mo
Included free
2K AI credits free (shared, worth $20)
Error analysis - 4 scoring dimensions, clustering
Insight generation - pattern detection across traces
Auto-tagging - AI-generated tags
Performance reports - AI-summarized analysis
Eval explanation - human-readable summaries
Pricing
First 2K credits Free
2K+ credits $10/1K
~10-30 credits per trace analysis. Uses shared AI Credits pool.
Calculator
Estimate your bill
Drag the sliders to see your estimated monthly cost.
Products
Tracing & Storage
First 50 GB free every month
Trace & eval storageTraces, spans, eval results, session data
$0/mo
050 GB2K
Platform add-ons
Boost
$250/mo
Scale
$750/mo
Enterprise
$2,000/mo
Estimated totalAll products + add-ons
$0/mo
Add-ons
Platform add-ons
Compliance, retention, and support. Subscribe in your account - no sales call.
Boost
$250 /mo
90-day data retention
5 knowledge bases
10 annotation queues · 15 monitors
SOC 2 Type II · OAuth SSO · audit logs
99.5% SLA · 48hr email support
Popular
Scale
$750 /mo
Everything in Boost, plus:
1-year retention · unlimited queues/monitors
Review workflow · inter-annotator agreement
HIPAA BAA · SAML SSO + SCIM
99.9% SLA · 24hr email · Slack channel
Enterprise
$2,000 /mo
Everything in Scale, plus:
Custom retention · ABAC · data masking
Dedicated support engineer + CSM
Training sessions · architecture review
Financial SLA · custom rate limits
Comparison
Compare alternatives
Feature-by-feature and dollar-for-dollar.
| Feature | Future AGI | Arize AI | LangSmith | Braintrust |
|---|---|---|---|---|
| Core Platform | ||||
| End-to-end tracing | ||||
| LLM evaluations | ||||
| Custom dashboards | AX only | Pro+ | ||
| Annotation queues + review | AX only | |||
| Gateway & Guardrails | ||||
| AI gateway (routing, caching, failover) | - | - | Basic proxy | |
| Real-time guardrails (18+ types) | - | - | - | |
| ML guardrails (Protect model) | - | - | - | |
| Evaluation & Testing | ||||
| Text + voice simulation | - | - | - | |
| Agentic evaluation (7 agents) | - | - | - | |
| BYOK evals ($0 platform cost) | ||||
| Managed eval models (Turing) | - | - | - | |
| Pricing & Plans | ||||
| No per-seat pricing | N/A | - | ||
| Unlimited seats on free | N/A | - | ||
| 50 GB free storage | 1 GB | 5K traces | 1 GB | |
| Self-hosted (MIT) | Elastic v2 | - | - | |
| SOC 2 Type II | $250/mo | Enterprise | Enterprise | Enterprise |
| HIPAA BAA | $750/mo | Enterprise | Enterprise | Enterprise |
What does it actually cost?
Same team, same usage. Future AGI includes Command Center (gateway), guardrails, and simulation - others charge extra or don't offer them.
| Scenario | Future AGI | Langfuse | LangSmith | Braintrust |
|---|---|---|---|---|
| Solo Developer 1 person, 10K traces/mo, basic evals | $0 | $0 | $13 | $0 |
| Seed Startup 5 people, 50K traces/mo, evals | $0 | $59 | $295 | $249 |
| Series A 15 people, 200K traces/mo, gateway, SOC 2 | $270 | $1,000+ ouch | $2,500+ yikes | $1,500+ oof |
| Series B 40 people, 1M traces/mo, HIPAA, 5M gateway | $2,200 | $4,000+ pain | $8,000+ wow | $5,000+ nope |
| Enterprise 100+ people, 5M traces/mo, full compliance | $7,000 | Call your CFO | Hope you raised | Good luck |
More included at every price point Future AGI pricing includes Command Center (AI gateway), 15+ guardrails, text & voice simulation, agentic evaluation, ML-powered Protect model, and unlimited team members - all included. LangSmith charges $39/seat + $2.50-5/1K traces. Braintrust Pro is $249/mo with only 5 GB + $3-4/GB overage. None of them include a gateway, guardrails, or simulation - you'd need to buy those separately.
Plans
Compare plans
Recommended
Free
No credit card required
- 50 GB storage/mo
- 2K AI credits/mo
- 100K gateway requests/mo
- 100K cache hits/mo
- 1M text sim tokens/mo
- 60 min voice sim/mo
- 30-day data retention
- Community support
Pay-as-you-go
Usage-based after free tier
Starts at $0 /mo
- Everything in Free
- Unlimited usage (pay-as-you-go)
- Volume discounts at scale
- 30-day data retention
- Email support
- Billing alerts & spending caps
Every plan includes
- Unlimited team members
- Unlimited projects
- Full API access
- All tracing & observability
- BYOK evals ($0 platform cost)
- 15 built-in gateway guardrails
- Datasets & annotations (forever)
- Experiments & prompt management
SOC 2, HIPAA, SSO, extended retention
FAQ
Questions & answers
How does usage-based pricing work?
There are 6 billing dimensions: Storage, AI Credits, Gateway Requests, Cache Hits, Text Simulation, and Voice Simulation. Each has a generous free tier that resets monthly. You only pay when you exceed those limits. Volume discounts apply automatically as you scale.
What happens when I exceed the free tier?
On the free plan, usage pauses at the cap - you never get a surprise bill. On pay-as-you-go (credit card required), usage continues seamlessly and you're billed for the overage at published rates with automatic volume discounts.
Do I need a credit card to start?
No. The free plan requires no credit card. You get 50 GB storage, 2K AI credits, 100K gateway requests, 100K cache hits, 1M text sim tokens, and 60 minutes of voice sim - every month. Add a card when you're ready to scale beyond those limits.
What are AI Credits?
AI Credits power our managed AI features: code evaluations and LLM-as-judge (charged per run), Protect guardrail checks (content moderation, PII, prompt injection), agentic evaluation agents, and synthetic data generation. Heuristic evals (BLEU, ROUGE, regex) are always free. You get 2K credits free/mo. Alternative: use BYOK (your own API key) for LLM-as-judge evals at $0 platform cost.
What's the difference between BYOK and AI Credits?
BYOK (Bring Your Own Key) means you provide your own OpenAI/Anthropic/etc. API key for evaluations - $0 platform cost, you pay the LLM provider directly. AI Credits use our managed Turing models - $10/1K credits, we handle the infrastructure. Most eval features work with either option.
What's included for free on all plans?
Unlimited team members, unlimited projects, full API access, tracing with 11 span types, 70+ trace filters, custom dashboards, heuristic evals, BYOK LLM-as-judge evals, datasets (stored forever), experiments, prompt management, annotation labels, agent playground, and 15 built-in gateway guardrails. No per-seat pricing - ever.
What do Boost, Scale, and Enterprise add?
Boost ($250/mo): 90-day data retention, 5 knowledge bases, 10 annotation queues, 15 monitors. Scale ($750/mo): 1-year retention, unlimited KBs/queues/monitors, review workflow, agreement metrics. Enterprise ($2,000/mo): SAML SSO, SCIM, HIPAA BAA, audit logs, ABAC, dedicated support, custom retention.
Can I set spending caps?
Yes. You can configure billing alerts at any threshold and set hard spending caps that pause ingestion when reached. The gateway has per-org, per-team, per-user, and per-key budget controls with hourly, daily, or monthly periods.
How does data retention work?
Free & Pay-as-you-go: 30-day retention. Boost: 90 days. Scale: 1 year. Enterprise: custom. Datasets and annotations persist forever on all plans. Gateway configs never expire.
Do you offer startup discounts?
Yes. Startups accepted into our program get $10K in free credits and 6 months of Scale-level access. Visit /startups or contact us for details.
Ready to ship AI
with confidence?
Start free with generous limits on every product. No credit card required.