text model comparison
GPT-6 Luna vs Hy4 flash
GPT-6 Luna and Hy4 flash are both current text models, from OpenAI and Tencent. Hy4 flash is 30% cheaper per 1M input tokens than GPT-6 Luna, $0.07 vs $0.10. In Tencent's Hy4 flash launch table, GPT-6 Luna scores 67.1% on DeepSWE v1.1 and Hy4 flash 64.6%. Both accept about 1.05M tokens of context (1,050,000 and 1,048,576).
- Lower price per 1M input tokens
- Hy4 flash$0.07 against $0.10
- Longer context window
- GPT-6 Luna1.05M tokens against 1.05M tokens
- Newer release
- Hy4 flashNov 2026 against Sep 2026
Data verified November 17, 2026 · Dataset 2026-11-17
Side by side
GPT-6 Luna and Hy4 flash, 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-6 Luna | Hy4 flash |
|---|---|---|
| Price | ||
| Price per 1M input tokens | $0.10 | $0.07 (better) |
| Price per 1M output tokens | $0.50 | $0.281 (better) |
| Independent benchmarks | ||
| Arena Text | 1441 Elo (#36 of 50)Range 1437 to 1445 | Not measured |
| Arena WebDev | 1577 Elo (#31 of 46)Range 1571 to 1583 | Not measured |
| Epoch Capabilities Index | 156.4 points (#25 of 43)Range 154.5 to 158.5 | Not measured |
| GPQA Diamond | 90.5% (#27 of 40) | Not measured |
| FrontierMath Tiers 1-3 (v2) | 78.9% (#15 of 39) | Not measured |
| SimpleQA Verified | 41.4% (#27 of 36) | Not measured |
| Specs | ||
| Provider | OpenAI | Tencent |
| Release date | September 22, 2026 | November 16, 2026 |
| Access | API only | API only |
| Context window | 1,050,000 tokens | 1,048,576 tokens |
| Max output | 128,000 tokens | 65,536 tokens |
| Input | text, image | text, image |
| Output | text | text |
Ranges under the scores are the publishers' own: published rating ranges from Arena and 90% credible intervals from Epoch AI. Where two ranges overlap, the gap is inside the measurement's own uncertainty, so the row says too close to call and neither score is highlighted or counted as a lead.
No independent benchmark has scored both models yet. Each rank counts the current models scored on that benchmark.
Provider results
What the labs report
Results tables a lab published with one of its models, where it lists both GPT-6 Luna and Hy4 flash. These are the lab's own figures, not independent measurements, and each table is read on its own: labs run benchmarks with different harnesses and budgets.
Tencent, Hy4 flash launch table, November 16, 2026Tencent · 2026-11-17
| Benchmark | GPT-6 Luna | Hy4 flash |
|---|---|---|
| DeepSWE v1.1 | 67.1% (better) | 64.6% |
| ProgramBenchalmost solved | 6.0% | 16.0% (better) |
| Terminal-Bench 4.0Claude Code harness | 13.3% | 19.2% (better) |
| APEX-Agents v1.1pass@1 | 44.7% | 63.1% (better) |
| AutomationBench v1.0.6 | 38.1% | 52.2% (better) |
| HLE-Fullwith tools | 49.7% | 55.2% (better) |
| OSWorld 2.1full, partial score | 55.7% | 68.1% (better) |
| MMMU-Pro | 77.5% (better) | 75.8% |
| ZeroBench mainwith tools, pass@5 | 33.6% | 45.8% (better) |
Tencent's appendix to the Hy4 flash announcement, every model at its highest available reasoning setting and run by Tencent. DeepSeek-V4.1-Flash is the 1017 build. Tencent's Hy4 table gives DeepSeek-V4.1-Flash 51.0% on ZeroBench, this one 16.5%. Left out here: Tencent's internal WorkIndex benchmarks, the cost per task in yuan and the Hy4 preview column.
Charts
Against every current text model
Where GPT-6 Luna and Hy4 flash land among the current models of their task, on every independent benchmark, on price and on context.
Where each stands
Only GPT-6 Luna has a current Arena Text score: #36 of 50. Each bar is the share of current models it beats.
Independent benchmarks
Arena Text
GPT-6 Luna:#36of 50
1,441 Elo · ahead of 29%
Published rating range 1,437 to 1,445
Hy4 flash: Not scored
Arena WebDev
GPT-6 Luna:#31of 46
1,577 Elo · ahead of 33%
Published rating range 1,571 to 1,583
Hy4 flash: Not scored
Epoch Capabilities Index
GPT-6 Luna:#25of 43
156.4 · ahead of 43%
90% credible interval 154.5 to 158.5
Hy4 flash: Not scored
GPQA Diamond
GPT-6 Luna:#27of 40
90.5% · ahead of 33%
Hy4 flash: Not scored
FrontierMath Tiers 1-3 (v2)
GPT-6 Luna:#15of 39
78.9% · ahead of 63%
Hy4 flash: Not scored
Price and context
Price per 1M input tokens
GPT-6 Luna:#2of 60
$0.10 · cheaper than 93%
Hy4 flash:#1of 60
$0.07 · cheaper than 100%
Price per 1M output tokens
GPT-6 Luna:#4of 60
$0.50 · cheaper than 90%
Hy4 flash:#2of 60
$0.281 · cheaper than 98%
Context window
GPT-6 Luna:#1of 71
1.05M tokens · longer than 86%
Hy4 flash:#12of 71
1.05M tokens · longer than 60%
Quality against price
GPT-6 Luna scores 1,441 at $0.2 (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 Carbon | $4 | 1,526 | |
| Gemini 4 Argon | $4 | 1,522 | |
| Muse Wonder | Meta | $8 | 1,511 |
| Claude Opus 5.5 | Anthropic | $8 | 1,507 |
| Gemini 3.8 Flash | $1.50 | 1,497 | |
| Muse Spark 1.3 | Meta | $2 | 1,497 |
| Gemini 4 Hydrogen | $3 | 1,490 | |
| Kimi K3 | Moonshot Kimi | $6 | 1,487 |
| GPT-5.6 Sol | OpenAI | $8 | 1,485 |
| Kimi K3.1 Pro | Moonshot Kimi | $4 | 1,485 |
| Qwen3.8-Max | Alibaba Qwen | $3 | 1,484 |
| Gemini 3.6 Flash | $3 | 1,483 | |
| GLM-5.3 | Z.ai | $2.15 | 1,482 |
| GPT-5.5 | OpenAI | $11 | 1,481 |
| GPT-6.1 Sol | OpenAI | $4 | 1,481 |
| MiMo-V2.6-Pro | Xiaomi MiMo | $0.54 | 1,480 |
| GPT-5.4 | OpenAI | $5.63 | 1,475 |
| DeepSeek-V4.1-Flash | DeepSeek | $0.52 | 1,475 |
| GLM-5.3-Flash | Z.ai | $0.24 | 1,475 |
| GPT-6 Astra | OpenAI | $20 | 1,474 |
| Gemini 3 Flash Preview | $1.13 | 1,473 | |
| Claude Sonnet 5.5 | Anthropic | $4 | 1,473 |
| Grok 4.20 | xAI | $1.56 | 1,472 |
| GPT-5.6 Terra | OpenAI | $4.50 | 1,468 |
| Grok 4.5 | xAI | $3 | 1,466 |
| DeepSeek-V4-Pro | DeepSeek | $1.98 | 1,464 |
| Grok 4.6 | xAI | $3 | 1,457 |
| GPT-6 Sol | OpenAI | $4 | 1,457 |
| Gemini 3.5 Flash-Lite | $0.85 | 1,455 | |
| GPT-5.6 Luna | OpenAI | $0.45 | 1,453 |
| Step 5 Preview | StepFun | $1.43 | 1,449 |
| MiMo-V2.6-Flash | Xiaomi MiMo | $0.18 | 1,448 |
| Grok 4.7 | xAI | $3 | 1,443 |
| Grok 4.3 | xAI | $1.56 | 1,442 |
| GPT-6 Luna | OpenAI | $0.20 | 1,441 |
| Claude Haiku 5.5 | Anthropic | $0.20 | 1,421 |
| Mistral Large 4 | Mistral AI | $1.05 | 1,420 |
| GPT-4.1 | OpenAI | $3.50 | 1,415 |
| Inkling 2 Small | Thinking Machines Lab | $0.47 | 1,412 |
| Solar Pro 4 | Upstage | $0.52 | 1,387 |
| 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-6 Luna costs 1.6 times as much as Hy4 flash, at each provider's listed rates.
10M input and 2M output tokens
GPT-6 Luna costs 1.6× as much
- GPT-6 Luna
- $2
- Hy4 flash
- $1.26
The same, with 9M of the input tokens cached
GPT-6 Luna costs 1.8× as much
- GPT-6 Luna
- $1.19
- Hy4 flash
- $0.66
| Workload or rate | GPT-6 Luna | Hy4 flash |
|---|---|---|
| Example workloads | ||
| 10M input and 2M output tokens | $2 | $1.26 (better) |
| The same, with 9M of the input tokens cached | $1.19 | $0.66 (better) |
| Other listed rates | ||
| 1M input tokens, cached | $0.01 | $0.003 (better) |
| 1M input tokens, prompts over 272K tokens | $0.20 | Not listed |
| 1M output tokens, prompts over 272K tokens | $0.75 | Not listed |
Your own bill will differ. These figures use list prices and example workloads. The AI Model Cost Audit reviews your real usage, shortlists models that score as well for each task for less, and estimates the savings.
Launch price $100, regular price $150
Verdict
When to choose each
GPT-6 Luna
Where it leads
- Longer responses, up to 128,000 vs 65,536 output tokens
Best for
- High-volume summarisation, extraction and routing
- Long-context budget work, where the 272k threshold is more forgiving than Haiku 5.5's 100k
Watch out for
- Prompts over 272,000 tokens are rebilled at $0.20 input and $0.75 output for the whole request
- A twentieth of GPT-6 Sol's price reflects a capability gap, not a discount on the same model
Hy4 flash
Where it leads
- Lower input and output prices, $0.07 and $0.281 vs $0.10 and $0.50 per 1M tokens
- Cheaper cached input, $0.003 vs $0.01 per 1M tokens
Best for
- High-volume everyday office work: deep search, documents, slides and simple front-end demos
- Cheap computer-use and agent steps where speed and cost per task matter more than peak accuracy
Watch out for
- In Tencent's own table it trails Hy4 on all nine public benchmarks and DeepSeek-V4.1-Flash on eight, most on Terminal-Bench 4.0 (19.2% against 44.4% and 40.0%)
- API only for now: Tencent plans open weights without giving a date, and recommends Hy4 instead for long-horizon software engineering
- No independent benchmark score is recorded here yet; Tencent's other evidence is internal: its WorkIndex and a blind test by 85 of its own WorkBuddy users, in which satisfaction came close to Claude Haiku 5.5
From benchmark to production
Building on GPT-6 Luna or Hy4 flash?
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.
AI Model Cost AuditLaunch price $100, regular price $150
FAQ
GPT-6 Luna vs Hy4 flash FAQ
Which is cheaper, GPT-6 Luna or Hy4 flash?
Hy4 flash is 30% cheaper per 1M input tokens than GPT-6 Luna, $0.07 vs $0.10. For 10M input and 2M output tokens, GPT-6 Luna costs $2 and Hy4 flash $1.26 at standard rates.
Which is cheaper with prompt caching, GPT-6 Luna or Hy4 flash?
GPT-6 Luna charges $0.01 per 1M cached input tokens, 90% below its standard input rate, and Hy4 flash $0.003, 96% below. With 9M of 10M input tokens cached and 2M output tokens, GPT-6 Luna costs $1.19 and Hy4 flash $0.66.
How do GPT-6 Luna and Hy4 flash compare in Tencent's own results?
In the Hy4 flash launch table (November 16, 2026), GPT-6 Luna leads on DeepSWE v1.1, 67.1% to 64.6%; Hy4 flash leads on ProgramBench (almost solved), 16.0% to 6.0%; Hy4 flash leads on Terminal-Bench 4.0 (Claude Code harness), 19.2% to 13.3%. These are Tencent's figures, not independent measurements.
Which has the larger context window, GPT-6 Luna or Hy4 flash?
Both accept about 1.05M tokens of context (1,050,000 and 1,048,576). GPT-6 Luna returns up to 128,000 tokens per response, Hy4 flash 65,536.
AI Model Brief
The AI models worth knowing, once a week
What shipped, how it scores on independent benchmarks, what it costs per million tokens and what was deprecated, from the data behind our AI model benchmarks.
We use your address only to send the brief. See our privacy policy.
See this week’s briefFurther reading
More on GPT-6 Luna and Hy4 flash
Articles on GPT-6 Luna and Hy4 flash
- OpenAI Models on Google Cloud: What Changes for AI TeamsOctober 22, 2026
- Claude Sonnet 5.5 vs GPT-6 Luna: Cost, Speed and AccuracySeptember 29, 2026
- GPT-6 Sol vs Claude Opus 5.5, Plus Luna: Best Value per TaskSeptember 24, 2026
- GPT-6 Sol and Luna: Half the Price, and Where Each One FitsSeptember 23, 2026
Related comparisons
- Claude Haiku 5.5 vs GPT-6 Luna
- DeepSeek-V4.1-Flash vs Hy4 flash
- GPT-6.1 Sol vs GPT-6 Luna
- GLM-5.3-Flash vs Hy4 flash
- Claude Sonnet 5.5 vs GPT-6 Luna
- Hy4 vs Hy4 flash
- Claude Opus 5.5 vs GPT-6 Luna
- DeepSeek-V4.1-Flash vs GPT-6 Luna
- GPT-5.4 vs GPT-6 Luna
- GPT-6 Astra vs GPT-6 Luna
- GPT-6 Luna vs GPT-6 Sol
- GPT-6 Luna vs Grok 4.3
Production evaluation
Choosing between GPT-6 Luna and Hy4 flash?
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
