Compare models side by side

PLATFORMPROMPTZDRPII STRIPPEDPrivacyWe provide end-to-end encryption and retain zero data. On Enterprise, we remove PII before prompts reach a closed model.TOK/SVERIFIEDSpeedArtificial Analysis verified us as the #1 Nemotron 3 Ultra provider at 454 tokens/sec, July 2026.
MODELS
Model catalogCompare modelsBlackbox Router
DEPLOYMENT
Dedicated deploymentSecurity overviewZero data retentionMicrosoft Azure
ENTERPRISE
Move faster with direct support from Blackbox engineers — from your first token to production scale.
TALK TO SALES →
PRODUCTSTENANT-02TENANT-03TENANT-04TENANT-05ISOLATED · SINGLE-TENANTREQUEST INTENANT-01RESERVED CAPACITY454 T/STOKENS OUTEnterprise InferenceSECURE INFERENCE FOR ENTERPRISEThe open-weight model that you choose, on capacity reserved for you — no shared pools, and no other customers on the instance.Blackbox Router300+ open and closed models through one endpoint. The gateway enforces zero data retention and no training, wherever the provider API supports it.Agents APISend coding agents to any repository over HTTP — create runs, stream logs, and open PRs.Command LineAn agentic terminal that reads the conventions of your repository.
RESOURCESGET /V1/USERS/{ID}API REFERENCESDK / GUIDESDeveloper DocumentationGUIDES · API REFERENCE · SDKSQuickstarts, the full API reference, and SDK guides — from your first key to production traffic.LATEST BLOGS
Announcing Nemotron 3.5 Lightning — reasoning at 1,200 tokens/secAUG 11, 2026 · 5 MIN READTB v2.1 Blackbox — GPT-5.6 Sol + Opus 4.8 (90% pass@1)JUL 21, 2026 · 2 MIN READBenchmark Performance: Faster Inference, Reference-Level Model QualityJUL 15, 2026 · 2 MIN READView all blogs →
COMPANY
MissionCareersPress kitBrand guide
GET STARTED
Create an accountLog in
TALK TO US
Questions about deployment, procurement, or a custom contract? We route you to a person, not a form reply.
CONTACT US →
PRICINGLOGINGET STARTED
MODEL CATALOG
Compare models.Side-by-side pricing, capacity, and capabilities for up to four models. Select models in the slots below. The URL updates as you select, so you can share any comparison.
BROWSE THE CATALOG
Models
openai/o3OpenAI
Empty
PROVIDER
OpenAI
—
TYPE
Text
—
CONTEXT WINDOW
200K
—
MAX OUTPUT
100K
—
INPUT PRICE
$2 / 1M tokens
—
OUTPUT PRICE
$8 / 1M tokens
—
REASONING
Yes
—
INPUT MODALITIES
file, image, text
—
ZERO DATA RETENTION
No
—
RELEASED
2025-04-16
—
TAGS
file-input, implicit-caching, reasoning, tool-use, vision, web-search
—
Suggested to add
Deep-link a comparison with ?ids=, e.g. /compare?ids=blackboxai/openai/o3
The high-trust inference layer for teams that deploy at scale.
BLACKBOX AI TECHNOLOGIES INC.
535 MISSION STREET
SAN FRANCISCO, CA
PLATFORM
Enterprise InferenceBlackbox RouterModel catalogNemotron 3 UltraCompare modelsAgents APIPricing
DEPLOYMENT
Security overviewDedicated deploymentZero data retentionMicrosoft AzureTalk to sales
DEVELOPERS
DocsAPI ReferenceCommand LineVS Code ExtensionLog in
COMPANY
MissionContact UsCareersBlogPress KitBrand Guide
© 2026 BLACKBOX AI TECHNOLOGIES INC. — ALL SYSTEMS OPERATIONALPRIVACY · TERMS