text model comparison
GPT-5.6 Luna vs Mercury 3
GPT-5.6 Luna and Mercury 3 are both current text models, from OpenAI and Inception. Both cost $0.20 per 1M input tokens. GPT-5.6 Luna has the larger context window, 1,050,000 vs 1,000,000 tokens.
- Longer context window
- GPT-5.6 Luna1.05M tokens against 1M tokens
- Newer release
- Mercury 3Oct 2026 against Jul 2026
Data verified October 23, 2026 · Dataset 2026-10-23
Side by side
GPT-5.6 Luna and Mercury 3, line by line
Prices in effect today, compared at the same tier wherever both list one. The better value in each row is highlighted.
| Fact | GPT-5.6 Luna | Mercury 3 |
|---|---|---|
| Price | ||
| Price per 1M input tokens | $0.20 | $0.20 |
| Price per 1M output tokens | $1.20 | $0.70 (better) |
| Independent benchmarks | ||
| Arena Text | 1452 Elo (#34 of 49) | Not measured |
| GPQA Diamond | 91.6% (#20 of 40) | Not measured |
| FrontierMath Tiers 1-3 (v2) | 82.1% (#12 of 40) | Not measured |
| SimpleQA Verified | 41% (#27 of 36) | Not measured |
| Specs | ||
| Provider | OpenAI | Inception |
| Release date | July 9, 2026 | October 19, 2026 |
| Access | API only | API only |
| Context window | 1,050,000 tokens | 1,000,000 tokens |
| Max output | 128,000 tokens | 131,072 tokens |
| Input | text, image | text |
| Output | text | text |
No independent benchmark has scored both models yet. Each rank counts the current models scored on that benchmark.
Charts
Against every current text model
Where GPT-5.6 Luna and Mercury 3 land among the current models of their task, on every independent benchmark, on price and on context.
Where each stands
Only GPT-5.6 Luna has a current Arena Text score: #34 of 49. Each bar is the share of current models it beats.
Independent benchmarks
Arena Text
GPT-5.6 Luna:#34of 49
1,452 Elo · ahead of 31%
Mercury 3: Not scored
GPQA Diamond
GPT-5.6 Luna:#20of 40
91.6% · ahead of 51%
Mercury 3: Not scored
FrontierMath Tiers 1-3 (v2)
GPT-5.6 Luna:#12of 40
82.1% · ahead of 72%
Mercury 3: Not scored
SimpleQA Verified
GPT-5.6 Luna:#27of 36
41% · ahead of 23%
Mercury 3: Not scored
Price and context
Price per 1M input tokens
GPT-5.6 Luna:#7of 49
$0.20 · cheaper than 83%
Mercury 3:#7of 49
$0.20 · cheaper than 83%
Price per 1M output tokens
GPT-5.6 Luna:#10of 49
$1.20 · cheaper than 77%
Mercury 3:#7of 49
$0.70 · cheaper than 88%
Context window
GPT-5.6 Luna:#1of 64
1.05M tokens · longer than 87%
Mercury 3:#26of 64
1M tokens · longer than 22%
Quality against price
GPT-5.6 Luna scores 1,452 at $0.45 (blended price per 1M tokens). The line joins the current models no other beats on both price and Arena Text score; hover any point to name it.
Quality against price: the data
| Model | Provider | Blended price per 1M tokens | Arena Text (Elo) |
|---|---|---|---|
| Gemini 4 Argon | $4 | 1,525 | |
| Claude Opus 5.5 | Anthropic | $8 | 1,507 |
| Claude Fable 5 | Anthropic | $20 | 1,504 |
| Claude Fable 5.1 | Anthropic | $20 | 1,501 |
| Gemini 3.8 Flash | $1.50 | 1,497 | |
| Muse Spark 1.3 | Meta | $2 | 1,494 |
| Claude Opus 5 | Anthropic | $10 | 1,490 |
| Kimi K3 | Moonshot Kimi | $6 | 1,488 |
| Gemini 3.7 Flash | $1.50 | 1,486 | |
| GPT-5.6 Sol | OpenAI | $8 | 1,485 |
| GPT-6.1 Sol | OpenAI | $4 | 1,484 |
| Qwen3.8-Max | Alibaba Qwen | $3 | 1,483 |
| GPT-5.5 | OpenAI | $11 | 1,482 |
| Gemini 3.6 Flash | $3 | 1,482 | |
| Claude Opus 4.8 | Anthropic | $10 | 1,481 |
| MiMo-V2.6-Pro | Xiaomi MiMo | $0.54 | 1,480 |
| GLM-5.3 | Z.ai | $2.15 | 1,478 |
| Claude Sonnet 5.5 | Anthropic | $4 | 1,476 |
| GPT-5.4 | OpenAI | $5.63 | 1,475 |
| GPT-6 Astra | OpenAI | $20 | 1,475 |
| DeepSeek-V4.1-Flash | DeepSeek | $0.52 | 1,475 |
| GLM-5.3-Flash | Z.ai | $0.24 | 1,475 |
| Gemini 3 Flash Preview | $1.13 | 1,473 | |
| Grok 4.20 | xAI | $1.56 | 1,472 |
| GPT-5.6 Terra | OpenAI | $4.50 | 1,466 |
| Grok 4.5 | xAI | $3 | 1,466 |
| DeepSeek-V4-Pro | DeepSeek | $1.98 | 1,465 |
| Claude Sonnet 5 | Anthropic | $4 | 1,461 |
| GPT-6 Sol | OpenAI | $4 | 1,456 |
| Gemini 3.5 Flash-Lite | $0.85 | 1,455 | |
| Grok 4.6 | xAI | $3 | 1,454 |
| GPT-5.6 Luna | OpenAI | $0.45 | 1,452 |
| MiMo-V2.6-Flash | Xiaomi MiMo | $0.18 | 1,451 |
| GPT-6 Luna | OpenAI | $0.20 | 1,443 |
| Grok 4.7 | xAI | $3 | 1,443 |
| Grok 4.3 | xAI | $1.56 | 1,442 |
| Mistral Large 4 | Mistral AI | $2.06 | 1,429 |
| GPT-4.1 | OpenAI | $3.50 | 1,415 |
| Solar Pro 4 | Upstage | $0.52 | 1,386 |
| GPT-4.1 mini | OpenAI | $0.70 | 1,383 |
| GPT-4.1 nano | OpenAI | $0.18 | 1,322 |
Cost
What each costs
Example workloads at each provider's listed rates, then every other rate either one lists. The cheaper figure in each row is highlighted.
The same work, priced
For 10M input and 2M output tokens, GPT-5.6 Luna costs 1.3 times as much as Mercury 3, at each provider's listed rates.
10M input and 2M output tokens
GPT-5.6 Luna costs 1.3× as much
- GPT-5.6 Luna
- $4.40
- Mercury 3
- $3.40
The same, with 9M of the input tokens cached
GPT-5.6 Luna costs 1.6× as much
- GPT-5.6 Luna
- $2.78
- Mercury 3
- $1.78
| Workload or rate | GPT-5.6 Luna | Mercury 3 |
|---|---|---|
| Example workloads | ||
| 10M input and 2M output tokens | $4.40 | $3.40 (better) |
| The same, with 9M of the input tokens cached | $2.78 | $1.78 (better) |
| Other listed rates | ||
| 1M input tokens, cached | $0.02 | $0.02 |
Verdict
When to choose each
GPT-5.6 Luna
Where it leads
- Larger context window 1,050,000 vs 1,000,000 tokens
- Accepts image input
Best for
- High-volume API evaluation with published low token prices
- Subagents and batch reasoning where cost per call dominates
Watch out for
- Not in the pinned LiveBench snapshot; its independent scores here come from Epoch AI's own GPQA Diamond, FrontierMath and SimpleQA Verified runs
Mercury 3
Where it leads
- Lower price per 1M output tokens, $0.70 vs $1.20
Best for
- Latency-critical agent loops and real-time search agents
- High-volume text generation at low cost
Watch out for
- The recorded rates are list prices; a 50% launch discount was running with no published end date
- Inception comparisons are its own, and no independent benchmark score is recorded here yet
From benchmark to production
Building on GPT-5.6 Luna or Mercury 3?
Public scores rarely hold on your own data. We test both models on your real tasks, then build the agent or pipeline around the one that wins.
FAQ
GPT-5.6 Luna vs Mercury 3 FAQ
Which is cheaper, GPT-5.6 Luna or Mercury 3?
Both cost $0.20 per 1M input tokens. For 10M input and 2M output tokens, GPT-5.6 Luna costs $4.40 and Mercury 3 $3.40 at standard rates.
Which is cheaper with prompt caching, GPT-5.6 Luna or Mercury 3?
GPT-5.6 Luna charges $0.02 per 1M cached input tokens, 90% below its standard input rate, and Mercury 3 $0.02, 90% below. With 9M of 10M input tokens cached and 2M output tokens, GPT-5.6 Luna costs $2.78 and Mercury 3 $1.78.
Which has the larger context window, GPT-5.6 Luna or Mercury 3?
GPT-5.6 Luna has the larger context window, 1,050,000 vs 1,000,000 tokens. Mercury 3 returns up to 131,072 tokens per response, GPT-5.6 Luna 128,000.
How do GPT-5.6 Luna and Mercury 3 differ in features?
GPT-5.6 Luna takes text and image input; Mercury 3 takes text.
Further reading
More on GPT-5.6 Luna and Mercury 3
Articles on GPT-5.6 Luna and Mercury 3
Related comparisons
- GPT-5.6 Luna vs GPT-5.6 Sol
- Gemini 3.1 Flash-Lite vs Mercury 3
- GPT-5.6 Luna vs GPT-5.6 Terra
- GLM-5.3-Flash vs Mercury 3
- GLM-5.3-Flash vs GPT-5.6 Luna
- Mercury 2.5 vs Mercury 3
- Gemini 3.1 Flash-Lite vs GPT-5.6 Luna
- Mercury 3 vs MiniMax M3.1 Flash
- GPT-5.6 Luna vs MiMo-V2.6-Flash
- GPT-5.6 Luna vs MiniMax M3.1 Flash
- GPT-5.6 Luna vs Qwen3.8-Omni-Flash
- GPT-5.6 Luna vs Mercury 2.5
Production evaluation
Choosing between GPT-5.6 Luna and Mercury 3?
Public benchmarks narrow the shortlist. We run both on your own tasks, data and cost limits, then build the routing and review controls to run the winner in production.
- A head-to-head evaluation on your real workload
- Routing between models by cost and difficulty