OpenAI · Efficient
GPT 6 Luna
GPT-6 Luna is the fast, cost-efficient model in OpenAI's GPT-6 series, suited for high-volume chat, classification, and lightweight agent workflows.
- Reasoning
- Tools
- Vision
- Structured output
- PDF input
- Input
- Text, Images
- Output
- Text
- Reasoning effort
- none · low · medium · high · xhigh · max
- Discovery tier
- Efficient
Pricing
USD per 1M tokens- Input
- $0.10
- Output
- $0.50
- Cache read
- $0.01
- Cache write
- $0.13
Reference prices as published by the model provider. API example
Specifications
- Context window
- 1.1M
- Max output
- 128K
- Average output (7 days)
- 120 tokens
- Longer than 12% of 473 models
Side by side
GPT 6 Luna compared
Specifications, prices, speed and benchmark scores next to each other. Benchmarks below show every compared model too.
| GPT 6 LunaOpenAIThis model | Gemini Pro LatestGoogleRemove | |
|---|---|---|
| Provider | OpenAI | |
| Discovery tier | Efficient | Frontier |
| Added | Sep 22, 2026 | Mar 29, 2026 |
| Limits | ||
| Context window | 1.05MBest | 1.05M |
| Max output | 128KBest | 65.54K |
| Average output | 120 tokens | — |
| Pricing, USD per 1M tokens | ||
| Input | $0.10Best | $2.00 |
| Output | $0.50Best | $12.00 |
| Cache read | $0.01Best | $0.20 |
| Cache write | $0.13Best | $0.38 |
| Performance | ||
| Throughput | 225.9 tok/sBest | 42.2 tok/s |
| Time to first token | 3.4 sBest | 5.1 s |
| Request success (24h) | 99.9% | 100.0%Best |
| Benchmark indexes | ||
| Intelligence Index | 29.9Best | 29.7 |
| Coding Index | — | 68.8 |
| Capabilities | ||
| Reasoning | Yes | Yes |
| Tools | Yes | Yes |
| Vision | Yes | Yes |
| Structured output | Yes | Yes |
| PDF input | Yes | No |
| Audio input | No | Yes |
| Video input | No | Yes |
| Reasoning effort | none, low, medium, high, xhigh, max | — |
Public evaluations
Benchmarks
Independent evaluation results, with how each score ranks among the models in the Noumen catalog.
- GPT 6 Luna
- Gemini Pro Latest
Intelligence Index
- GPT 6 Luna29.9
- Gemini Pro Latest29.7
GPT 6 Luna tested at medium reasoning effort
GPT 6 Luna: Better than 73% of 326 scored models
Indexes
Coding Index
Composite of the coding evaluations
- GPT 6 Luna: —
- Gemini Pro Latest: 68.8
Agentic work
AutomationBench-AA
Workflow automation, penalized for guardrail violations
- GPT 6 Luna: 40.5%Top
- Gemini Pro Latest: 35.4%
AutomationBench-AA tasks completed
Workflows finished end to end without a guardrail violation
- GPT 6 Luna: 10.2%Top
- Gemini Pro Latest: 9.4%
Harvey LAB-AAv1.1
Legal tasks passed in full, with no material hallucination
- GPT 6 Luna: —
- Gemini Pro Latest: 0.0%
AA-Briefcasev1.1
Agentic knowledge work, rated head to head
- GPT 6 Luna: 1,097 EloTop
- Gemini Pro Latest: 456 Elo
GDPval-AAv2.1
Economically valuable occupational tasks, rated head to head
- GPT 6 Luna: 1,263 EloTop
- Gemini Pro Latest: 791 Elo
Document reasoning
GDP.pdf
Professional reasoning over PDF documents, all-pass rate
- GPT 6 Luna: 13.8%
- Gemini Pro Latest: 17.8%Top
AA-LCRv1.1
Reasoning across long documents
- GPT 6 Luna: 78.3%
- Gemini Pro Latest: 82.0%Top
MLCR-AA
Long-context reasoning over medical records
- GPT 6 Luna: 18.3%Top
- Gemini Pro Latest: 11.1%
Reasoning and knowledge
Humanity’s Last Exam
Expert-level questions across academic fields
- GPT 6 Luna: 28.3%
- Gemini Pro Latest: 47.0%Top
GPQA Diamond
Graduate-level science questions
- GPT 6 Luna: —
- Gemini Pro Latest: 94.1%
IFBench
Following precise output instructions
- GPT 6 Luna: —
- Gemini Pro Latest: 77.1%
Coding
Terminal-Benchv4.0
Practical software engineering in a terminal
- GPT 6 Luna: 2.5%
- Gemini Pro Latest: 4.0%Top
SciCode
Scientific computing in Python
- GPT 6 Luna: 50.9%
- Gemini Pro Latest: 58.7%Top
Legacy evaluations
Terminal-Bench Hard
Earlier terminal task suite
- GPT 6 Luna: —
- Gemini Pro Latest: 53.8%
τ²-Bench Telecom
Tool use in a customer-support simulation
- GPT 6 Luna: —
- Gemini Pro Latest: 95.6%
AA-LCR (unversioned)
Long-context reasoning, earlier grading
- GPT 6 Luna: 78.3%
- Gemini Pro Latest: 82.0%Top
Scores from Artificial Analysis, updated Oct 26, 2026. GPT 6 Luna was tested at medium reasoning effort; compare scores measured at the same effort.Artificial Analysis
- GPT 6 Luna
- Gemini Pro Latest
| Leaderboard | GPT 6 Luna | Gemini Pro Latest |
|---|---|---|
| ARC-AGI-3 (Default harness) | #18 · 0.2% | Not in top 20 |
Ranks among Noumen catalog models on ARC Prize leaderboards. Only top-20 placements are listed.ARC Prize
- GPT 6 Luna
- Gemini Pro Latest
| Leaderboard | GPT 6 Luna | Gemini Pro Latest |
|---|---|---|
| SVG | #16 · 1250 | Not in top 20 |
Ranks among Noumen catalog models on Design Arena leaderboards. Only top-20 placements are listed.Design Arena
Live routing data
Performance and reliability
Recent averages across requests routed for this model. Your own speed depends on prompt length, reasoning effort and load.
- Throughput
- 225.9 tok/s
- Faster than 87% of 260 models
- Time to first token
- 3.4 s
- Quicker than 47% of 260 models
- Prompt cache hits
- 54.0%
- About $0.0514 per 1M input tokens with caching
- Request success (24h)
- 99.9%
- Reliable
Cache hits are the share of input tokens recently served from a prompt cache across all requests. They are not a forecast of your own hit rate.
- Healthy (98% or more)
- Degraded (90–98%)
- Down (under 90%)
- No requests
Developers
Use it through the API
Noumen’s API is OpenAI-compatible: any Chat Completions client works with a base URL and a key.
curl https://www.noumen.xyz/api/v1/chat/completions \
-H "Authorization: Bearer $NOUMEN_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "openai/gpt-6-luna",
"reasoning_effort": "medium",
"messages": [{ "role": "user", "content": "Explain this stack trace." }]
}'import os
from openai import OpenAI
client = OpenAI(base_url="https://www.noumen.xyz/api/v1", api_key=os.environ["NOUMEN_API_KEY"])
reply = client.chat.completions.create(
model="openai/gpt-6-luna",
messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)OpenAI
Related models
Other OpenAI models, starting with the same family.