Cyberax AI Playbook
cyberax.com
← All models
Model · OpenAI · Text

GPT-6 Luna

OpenAI's most efficient model for focused, high-volume tasks — the cheap tier of the GPT-6 generation. OpenAI's named replacement for GPT-4o, GPT-4o mini, GPT-4.1, GPT-4.1 mini and GPT-5.4 nano.

Modality
Text
License
Proprietary (Proprietary)
Context window
1,050,000 tokens 1.05M-token context (922k max input) with 128k max output. Knowledge cutoff 2026-05-18. Prompts over 272k input tokens cost 2x input and cache rates and 1.5x output for the whole request. Reasoning effort runs none / low / medium (default) / high / xhigh / max.
Released
September 22, 2026
Last verified
October 28, 2026
Runs locally
No
Also handles
vision, reasoning, code

Strengths

  • Lowest price in the GPT-6 generation ($0.10 input / $0.50 output per MTok; $0.01 cached input)
  • 1.05M-token context with 128k max output and a 2026-05-18 knowledge cutoff
  • Full hosted tool set in the Responses API — web search, file search, image generation, code interpreter, hosted shell, apply patch, skills, computer use, MCP, and tool search
  • Supports the `none` reasoning effort for the cheapest, fastest replies; Batch and Flex at 50% off
  • The model behind OpenAI's Decisions API (generally available 2026-10-22)

Weaknesses

  • Long prompts cost more — over 272k input tokens the whole request is billed at 2x input and 1.5x output
  • Chat Completions supports function calling only at `reasoning_effort: none`; built-in tools need the Responses API
  • Text and image input only — no audio or video
  • Not available for Realtime, Assistants, fine-tuning, embeddings, image generation, speech, transcription, or moderation
  • Closed weights — API and product only

Try it

WhereTypeNotes
OpenAI API hosted-api API key required

Used in solutions

How-to

Ad creative A/B testing at scale
How-to

AI agents for inbound qualification
How-to

Audit-trail generation from system logs
How-to

Auto-generate documentation from PRs and code
How-to

Auto-tag and route inbound social DMs
How-to

Automated invoice and receipt processing
How-to

Brand-voice guardrails for marketing teams
How-to

Detect churn signal from support patterns
How-to

Competitor monitoring with automated alerts
How-to

Compliance evidence collection for SOC 2 / ISO 27001
How-to

Content performance attribution
How-to

Contract review and clause extraction
How-to

CRM data hygiene at scale
How-to

Customer health scoring from product and support signals
How-to

Customer testimonial mining from reviews and support
How-to

Data entry automation from scanned forms
How-to

Document classification at scale
How-to

Email-to-task automation
How-to

Expense report categorization and anomaly detection
How-to

Federated search across your tools
How-to

Generate FAQ content from existing docs
Comparison

GPT vs Claude vs Gemini for business writing
How-to

Hook generation for short-form video
How-to

Insurance policy portfolio review
How-to

Internal Q&A bot over company docs
How-to

Investor updates from BI data
How-to

Lease and vendor renewal tracking
How-to

Meeting summaries people actually read
How-to

Multilingual customer support routing
How-to

Newsletter automation from weekly content
How-to

Onboarding documentation generation for new hires
Explainer Local-OK

Open-source vs proprietary AI — practical tradeoffs
How-to

Outbound prospecting research at SDR scale
How-to

Personalised email at 1-to-1000 scale
How-to

Procurement RFP response comparison
How-to

Programmatic SEO at scale
How-to

Quote-to-cash automation
How-to

Reactivation campaigns for dormant accounts
How-to

Reply templates learning from past conversations
How-to

Resume screening with anti-bias guardrails
How-to

Sales-call coaching at scale
How-to

Sales follow-up sequences with CRM context
How-to

SEO content audit at scale
How-to

Slack channel summaries that catch what matters
How-to

Social listening and brand-mention triage
How-to

SOP extraction from interviews
How-to

Tax document classification and extraction
How-to

Triage inbound email at scale
How-to

Vendor risk assessment from questionnaires
How-to

Voice-of-customer reports from cross-channel feedback

Version history

  1. GPT-6.1 Terra Oct 2026
  2. GPT-6.1 Sol Sep 2026
  3. GPT-6 Luna Sep 2026 Current
  4. GPT-6 Astra Sep 2026

Official sources

Change log

  • — Initial entry. Released 2026-09-22 alongside GPT-6 Sol at $0.10 input / $0.50 output per MTok, with a 1.05M-token context and 128k max output. Named on 2026-10-14 as the replacement for GPT-4o, GPT-4o mini, GPT-4.1 and GPT-4.1 mini (shutdown 2027-01-14).

Related models

GPT-6.1 Terra

OpenAI · Text

A fast, efficient GPT-6 model for visual reasoning, precise computer use, and clear, natural writing. Priced between GPT-6 Luna and GPT-6.1 Sol.

Proprietary 1.1M ctx
Try on: OpenAI API
Released Oct 2026 · Verified 2026-10-28

MiniMax M3.1 Flash

MiniMax · Text

MiniMax's low-cost model for agents, coding and computer use, at $0.15 in / $0.60 out per MTok. API-only for now; MiniMax says the weights will be released later.

Proprietary 1M ctx
Try on: MiniMax API
Released Oct 2026 · Verified 2026-10-28

Nemotron 3.5 Bolt 74B-A9B

NVIDIA · Text

NVIDIA's local agentic model, packed into a 24GB download so it runs on a single high-end consumer GPU (32 GB+) — text, image and video input with a 256k context. A compressed version of the 120B-A12B Nemotron 3.5 Super VL, NVIDIA's open data-centre model released the same day.

OpenMDW-1.1 Open weights Runs locally Mac · 32 GB+ 262k ctx
Try on: Hugging Face · Ollama / llama.cpp / LM Studio
Released Oct 2026 · Verified 2026-10-28

Gemini 4 Argon

Google · Text

Google's flagship reasoning model and the first of the Gemini 4 generation — its most capable model for complex coding, agentic, math, and science work. Takes over the Pro tier from Gemini 3.1 Pro Preview.

Proprietary 1.0M ctx
Try on: Google AI Studio · Vertex AI
Released Oct 2026 · Verified 2026-10-28

Claude Haiku 5.5

Anthropic · Text

Anthropic's fastest, cheapest tier — built for high-volume, latency-sensitive work such as classification, extraction, routing, and subagent tasks. Succeeds Claude Haiku 4.5.

Proprietary 1M ctx
Try on: Anthropic API · AWS Bedrock · Google Vertex AI
Released Oct 2026 · Verified 2026-10-28

Mistral Large 4

Mistral AI · Text

Mistral AI's new flagship (v26.10, API ID mistral-large-4) — a general-purpose multimodal model that Mistral describes as open-weight, with weights not yet published. Succeeds Mistral Large 3. EU-jurisdiction provider.

Proprietary 1M ctx
Try on: Mistral La Plateforme
Released Oct 2026 · Verified 2026-10-28