Punches far above its price: GPQA Diamond 92.3%, SWE-bench Pro 62.7%, Terminal-Bench 2.1 84.7%
$0.20/$1.20 per 1M after the July 30, 2026 price cut — dramatically cheaper per token than Gemini 3.6 Flash
Full 1.05M-token context at budget pricing — larger than most rival small models
Weaknesses
Long-context recall collapses at scale: 41.3% on 512K–1M token tasks vs Terra's 72.5%
Text and image input only — no video, audio, or native PDF ingestion like Gemini 3.6 Flash
Real-world use cases
What people actually use GPT-5.6 Luna for.
High-volume summarization and drafting at $0.20/1M input
Routine steps in agent pipelines where Sol/Terra would be overkill
Budget coding assistance — 62.7% SWE-bench Pro within ~2 points of Sol at 1/25th the output cost
How GPT-5.6 Luna compares
The nearest models people weigh against it, and what actually separates them.
vs GPT-6 Luna — Against GPT-6 Luna (OpenAI), GPT-5.6 Luna costs about 57% more per token and answers slower. GPT-6 Luna is the one to check first if the price difference matters more than the ceiling.
vs Claude Haiku 5.5 — Against Claude Haiku 5.5 (Anthropic), GPT-5.6 Luna costs about 57% more per token, takes 1.1x the context and answers slower. Claude Haiku 5.5 is the one to check first if the price difference matters more than the ceiling.
vs GPT-3.5 Turbo — Against GPT-3.5 Turbo (OpenAI), GPT-5.6 Luna runs about 30% cheaper per token, takes 64.1x the context and answers slower. Which one wins depends on whether context depth or latency is your constraint.
Price History
GPT-5.6 Luna pricing over time
↑100% since Aug 7
86 data points · tracked daily since Aug 7, 2026
Ready to try it?
Start using GPT-5.6 Luna
Cheap high-throughput summarization, drafting, and routine agent steps. Start free — no card required.
OpenAI's September 22, 2026 low-cost reasoning model, replacing GPT-5.6 Luna at $0.10/$0.50 per 1M — half the predecessor's price, with the same 1.05M context.
Verdict
OpenAI's budget tier at half the old price.
Quality score
80%
Pricing
$0.10/1M in
$0.50/1M out
Speed
Very fast
5/5 speed
Context
1.1M tokens
Released September 22, 2026 with GPT-6 Sol. Gateway id openai/gpt-6-luna. $0.10/$0.50 per 1M up to 272K input tokens, $0.20/$0.75 above; Flex $0.05/$0.25. 1,050,000 context, 128,000 max output, knowledge cutoff May 18, 2026. Verified October 10, 2026.
BudgetFastHigh volumeNew
Best for
High-volume, focused tasks where cost per call decides the model
Anthropic's October 7, 2026 low-cost model and the first Haiku with adjustable reasoning effort — thinking can be switched off for simple requests or raised to max, at $0.10/$0.50 per 1M under 100K input tokens.
Verdict
Anthropic's budget model, now with adjustable reasoning, at $0.10/$0.50.
Quality score
82%
Pricing
$0.10/1M in
$0.50/1M out
Speed
Very fast
5/5 speed
Context
1M tokens
Announced September 28, 2026 alongside Sonnet 5.5; generally available October 7, 2026. Gateway id anthropic/claude-haiku-5.5. Tiered pricing: $0.10/$0.50 per 1M up to 100K input tokens, $0.50/$2.50 above. 1,000,000 context, 128,000 max output. Launch figures as reported by Anthropic: Terminal-Bench 4.0 39.2 (max effort), 31.6 (extra), 12.8 (low); OSWorld 2.1 offline subset 72.4 (Haiku 4.5: 15.7). Verified October 10, 2026.
BudgetFastHigh volumeSub-agentsNew
Best for
High-volume tool use, sub-agents and everyday tasks on a tight budget
GPT-3.5 Turbo is OpenAI's legacy fast and affordable chat model, optimized for dialogue and straightforward text tasks at low cost. It was the backbone of early ChatGPT and remains a go-to for high-volume, cost-sensitive deployments.
Verdict
A once-dominant budget model now outclassed by cheaper, smarter alternatives like GPT-4o mini.
Quality score
35%
Pricing
$0.50/1M in
$1.50/1M out
Speed
Very fast
5/5 speed
Context
16k tokens
GPT-3.5 Turbo is still available via OpenAI API and supports fine-tuning, which keeps it relevant for teams with existing trained models. However, OpenAI has deprioritized its development in favor of the GPT-4o family. Not multimodal — text only.
BudgetLegacyFastHigh-volumeChatbot
Best for
High-volume, low-complexity tasks like chatbots, classification, summarization, and simple Q&A where cost matters more than cutting-edge quality.
GPT-5.6 Luna costs $0.2 per million input tokens and $1.2 per million output tokens on the API, with cached input at $0.02 per million. A month of 10M input and 2M output tokens runs about $4.40 at list price, before any batch or caching discounts.
What is the context window of GPT-5.6 Luna?
GPT-5.6 Luna has a 1.1M tokens context window, with up to 128k tokens of output per response. That is the total of prompt plus response the model can hold in one request.
What is the knowledge cutoff of GPT-5.6 Luna?
GPT-5.6 Luna's training data runs through February 16, 2026, and the model was released on July 9, 2026. For anything after that date it needs web search or documents in the prompt.
What is GPT-5.6 Luna best for?
GPT-5.6 Luna is best for cheap high-throughput summarization, drafting, and routine agent steps. It is a strong fit when that workflow matters more than the tradeoffs around budget pricing and fast speed.
When should I avoid GPT-5.6 Luna?
Your workload actually uses the long context window — recall drops to 41% past 512K tokens.
What is a cheaper alternative to GPT-5.6 Luna?
Claude Haiku 5.5 (Anthropic) at $0.10/1M/1M input against GPT-5.6 Luna's $0.20/1M/1M — roughly 57% less per token all in. Anthropic's budget model, now with adjustable reasoning, at $0.10/$0.50. Compare it first if GPT-5.6 Luna's pricing is the thing stopping you.
What is a faster alternative to GPT-5.6 Luna?
GPT-6 Luna — very fast against GPT-5.6 Luna's fast, with 1.1M tokens of context. Worth the swap when response time is what your users notice rather than the last few points of reasoning depth.
Newsletter
Get notified when GPT-5.6 Luna pricing changes
We track pricing daily. When this model drops or spikes, you'll know first.
No spam. Useful updates only. Affiliate disclosures always clearly labeled.