GPT-5.6 Luna

OpenAI has cut the price of GPT-5.6 Luna by 80% and GPT-5.6 Terra by 20% in a July 31, 2026 move that resets the price-performance bar for the family. Reporting from InfotechLead ties the reduction to a broader market trend in which falling inference costs are enabling wider deployment of agentic and high-volume worklo For Venice-hosted GPT-5.6 Luna users accessing the model via the OpenAI-compatible API at https://api.venice.ai/api/v1, the cut directly translates into lower token costs for production traffic and high-volume batch jobs, making Luna more attractive as a default low-cost frontier option. The story pairs Luna's aggressi

OpenAI announced on July 30, 2026 that GPT-5.6 Luna, its fastest and most affordable model in the GPT-5.6 family, will cost 80% less, with the lower price also reflected in how usage is counted against paid subscriptions in Codex and ChatGPT Work. The post positions Luna as delivering frontier-level intelligence on the The pricing and packaging changes are framed by OpenAI as the result of efficiency gains across every layer of how GPT-5.6 is built, served, and put to work, with the company stating that making advanced intelligence more abundant and affordable is central to its mission. The post explicitly names GPT-5.6 Luna as the c

OpenAI has expanded access to its GPT-5.6 model family beyond the limited preview announced last month, making the three variants—GPT-5.6 Sol, GPT-5.6 Terra, and GPT-5.6 Luna—generally available. Sol is positioned as the flagship for complex reasoning, coding, scientific work, cybersecurity, and long-running agentic ta The release introduces a new "max reasoning effort" setting for Sol and an "ultra mode" that coordinates subagents to accelerate complex reasoning. Reported benchmark highlights include Agents' Last Exam scores of 53.6 for Sol (13.1 points above Claude Fable 5), a near-parity Artificial Analysis Intelligence Index resu

LLM-Stats published a profile page for GPT-5.6 Luna with a composite LLM Stats Score of 45.1 at a blended price of $0.25 per million tokens, ranking it 32nd overall on the leaderboard. Capability tier breakdowns place Luna in the top 10% for Coding (21 of 267) and Reasoning (36 of 363), average in the top half for Webs The page situates GPT-5.6 Luna on a cost-efficiency scatter alongside Gemma 4 E4B ($0.024, score 13.8), GPT OSS 120B ($0.043, 28.8), DeepSeek-V4-Flash-0731 ($0.066, 44.7), DeepSeek-V4.1-Flash ($0.24, 51.8), and GPT-6 Astra ($11.9, 59.6), and lists per-benchmark results including a perfect 1.00/1 score on OpenAI's inter

VentureBeat reported on June 26, 2026 that OpenAI announced a limited preview of the GPT-5.6 family in three variants, with Sol positioned for the hardest problems such as complex coding and security research, Terra for high-volume business tasks like customer support, internal tools, and document analysis, and Luna fo The staggered release is contextualized as following a June 2, 2026 executive order calling on federal agencies to collaborate on a process for benchmarking and assessing new AI models before wide release, with the order said to take 30 days (until July 2, 2026). The piece also notes the contemporaneous U.S. export con

OpenAI began a limited preview of the GPT-5.6 series on June 26, 2026, introducing three variants: Sol as the flagship model, Terra as a balanced model for everyday work, and Luna as a fast and affordable model. The post states that Terra has competitive performance to GPT-5.5 while being 2x cheaper, and Luna brings st The launch post frames the staged rollout as a short-term step taken at the U.S. government's request, coordinated with the Administration on a cyber Executive Order framework and a repeatable process for future model releases, and states that OpenAI does not believe this kind of government access process should become

Aggregated OpenAI release notes dated July 31, 2026 confirm that GPT-5.6 Luna is now priced at $0.20 per million input tokens and $1.20 per million output tokens following an 80% reduction, with Terra moving to $2 and $12 per million in/out respectively after a 20% cut. The same announcement introduces a Fast mode for For developers reaching Luna through Venice's OpenAI-compatible endpoint (model slug openai-gpt-56-luna, base URL https://api.venice.ai/api/v1), the new pricing makes Luna one of the cheapest frontier-tier options available on the platform and reinforces its positioning as the high-volume, lowest-cost tier of the GPT-5