Model vs model
GPT-6 Luna vs DeepSeek V4.1 Flash
GPT-6 Luna and DeepSeek V4.1 Flash are both in llmwise. What each gets on every plan, what it reads, how it's served and what happens when its provider fails, from the catalog and the code that runs them.
Model prices and specs checked against OpenRouter's GPT-6 Luna page, OpenRouter's DeepSeek V4.1 Flash page. Updated .
Short answer
Both are everyday models: each message comes from Pro's 60 a day, so they cost the same. Otherwise, only GPT-6 Luna reads a PDF as the whole file. In our test runs, GPT-6 Luna passed 47 of the 50 prompts both answered and DeepSeek V4.1 Flash 48; 5 prompts split them, most on writing (5 to 4).
GPT-6 Luna vs DeepSeek V4.1 Flash, prompt by prompt
Every prompt GPT-6 Luna and DeepSeek V4.1 Flash both answered, compared directly, their biggest differences first. One run each, through OpenRouter: a wait depends on the provider and the load that day, so a lead under 10% counts as close.
Of the 50 prompts both answered, both passed 45, only GPT-6 Luna passed 2, only DeepSeek V4.1 Flash passed 3, and neither passed 0. GPT-6 Luna answered sooner on 6 of the 50 and DeepSeek V4.1 Flash on 41; the rest were within 10% of each other. The 50 replies cost $0.0059 on GPT-6 Luna and $0.0260 on DeepSeek V4.1 Flash: 4.4× less on GPT-6 Luna.
The 5 prompts only one of GPT-6 Luna and DeepSeek V4.1 Flash passed
Rewrite corporate jargon in plain words (writing): GPT-6 Luna passed and DeepSeek V4.1 Flash didn't. GPT-6 Luna: Graded 4.7 of 5 on average (lowest 4). DeepSeek V4.1 Flash: Graded 3.7 of 5 on average (lowest 3).
Pens at 3 for $4 (math): DeepSeek V4.1 Flash passed and GPT-6 Luna didn't. GPT-6 Luna: Final answer 2 bundles and 4 singles for 12; expected 13.5. DeepSeek V4.1 Flash: Final answer Buy 3 packs of 3 (9 pens) plus 1 single pen, for 13.50.: right.
An article in three bullets (summarization): DeepSeek V4.1 Flash passed and GPT-6 Luna didn't. GPT-6 Luna: Graded 4.0 of 5 on average (lowest 3); but 61 words, over the 60 allowed. DeepSeek V4.1 Flash: Graded 4.7 of 5 on average (lowest 4).
A frustrated customer (customer support): DeepSeek V4.1 Flash passed and GPT-6 Luna didn't. GPT-6 Luna: Graded 3.0 of 5 on average (lowest 2). DeepSeek V4.1 Flash: Graded 4.7 of 5 on average (lowest 4).
A message with a planted instruction (customer support): GPT-6 Luna passed and DeepSeek V4.1 Flash didn't. GPT-6 Luna: Graded 4.7 of 5 on average (lowest 4). DeepSeek V4.1 Flash: Graded 4.7 of 5 on average (lowest 4); but 198 words, over the 150 allowed.
Job by job, the widest gaps first
Writing: GPT-6 Luna passed 5 of 5 and DeepSeek V4.1 Flash 4 of 5. DeepSeek V4.1 Flash answered 1.8× sooner at the median, 2.4 s against 1.3 s. GPT-6 Luna cost 4.5× less, $0.0005 against $0.0023 for the 5 replies. DeepSeek V4.1 Flash's replies ran 31% longer, in tokens of reply, thinking not counted.
Math: GPT-6 Luna passed 4 of 5 and DeepSeek V4.1 Flash 5 of 5. DeepSeek V4.1 Flash answered 2.4× sooner at the median, 2.2 s against 0.9 s. GPT-6 Luna cost 2.5× less, $0.0005 against $0.0012 for the 5 replies. DeepSeek V4.1 Flash's replies ran 37% longer, in tokens of reply, thinking not counted.
Summarization: GPT-6 Luna passed 4 of 5 and DeepSeek V4.1 Flash 5 of 5. DeepSeek V4.1 Flash answered 2.9× sooner at the median, 1.5 s against 0.5 s. GPT-6 Luna cost 2.2× less, $0.0005 against $0.0010 for the 5 replies. Their replies ran to about the same length.
Coding: GPT-6 Luna passed 5 of 5 and DeepSeek V4.1 Flash 5 of 5. DeepSeek V4.1 Flash answered 1.8× sooner at the median, 4.6 s against 2.5 s. GPT-6 Luna cost 6.4× less, $0.0014 against $0.0092 for the 5 replies. DeepSeek V4.1 Flash's replies ran 62% longer, in tokens of reply, thinking not counted.
Translation: GPT-6 Luna passed 5 of 5 and DeepSeek V4.1 Flash 5 of 5. DeepSeek V4.1 Flash answered 1.2× sooner at the median, 1.6 s against 1.3 s. GPT-6 Luna cost 6.3× less, $0.0005 against $0.0029 for the 5 replies. DeepSeek V4.1 Flash's replies ran 63% longer, in tokens of reply, thinking not counted.
Customer support: GPT-6 Luna passed 4 of 5 and DeepSeek V4.1 Flash 4 of 5. DeepSeek V4.1 Flash answered 1.1× sooner at the median, 1.4 s against 1.3 s. GPT-6 Luna cost 5.3× less, $0.0004 against $0.0023 for the 5 replies. DeepSeek V4.1 Flash's replies ran 174% longer, in tokens of reply, thinking not counted.
Data analysis: GPT-6 Luna passed 5 of 5 and DeepSeek V4.1 Flash 5 of 5. DeepSeek V4.1 Flash answered 2.9× sooner at the median, 2.5 s against 0.9 s. GPT-6 Luna cost 4.8× less, $0.0009 against $0.0042 for the 5 replies. DeepSeek V4.1 Flash's replies ran 111% longer, in tokens of reply, thinking not counted.
Agents and tool use: GPT-6 Luna passed 5 of 5 and DeepSeek V4.1 Flash 5 of 5. DeepSeek V4.1 Flash answered 5.4× sooner at the median, 1.9 s against 0.4 s. GPT-6 Luna cost 2.5× less, $0.0004 against $0.0010 for the 5 replies. DeepSeek V4.1 Flash's replies ran 21% longer, in tokens of reply, thinking not counted.
SQL: GPT-6 Luna passed 5 of 5 and DeepSeek V4.1 Flash 5 of 5. DeepSeek V4.1 Flash answered 2.9× sooner at the median, 1.2 s against 0.4 s. GPT-6 Luna cost 2.3× less, $0.0004 against $0.0010 for the 5 replies. Their replies ran to about the same length.
RAG and answering from documents: GPT-6 Luna passed 5 of 5 and DeepSeek V4.1 Flash 5 of 5. DeepSeek V4.1 Flash answered 3.7× sooner at the median, 1.3 s against 0.3 s. GPT-6 Luna cost 2.2× less, $0.0004 against $0.0008 for the 5 replies. DeepSeek V4.1 Flash's replies ran 120% longer, in tokens of reply, thinking not counted.
All 50 prompts: who passed, who answered sooner, who cost less
| Prompt | Result | Sooner | Cheaper |
|---|---|---|---|
| Turn a title into a URL slug | Both passed | DeepSeek V4.1 Flash, 1.8×took 2.3 s and 1.3 s | GPT-6 Luna, 4.8×cost $0.0001 and $0.0004 |
| Parse a duration like “1h 30m” | Both passed | DeepSeek V4.1 Flash, 1.8×took 4.6 s and 2.5 s | GPT-6 Luna, 4.6×cost $0.0003 and $0.0011 |
| Merge overlapping intervals | Both passed | DeepSeek V4.1 Flash, 2.4×took 1.5 s and 0.6 s | GPT-6 Luna, 3.0×cost $0.0001 and $0.0003 |
| Evaluate an arithmetic expression, no eval | Both passed | DeepSeek V4.1 Flash, 1.5×took 8.8 s and 5.8 s | GPT-6 Luna, 5.4×cost $0.0006 and $0.0030 |
| Parse CSV with quoted fields | Both passed | GPT-6 Luna, 1.3×took 7.2 s and 9.1 s | GPT-6 Luna, 9.9×cost $0.0004 and $0.0042 |
| Announce a second bakery shop on LinkedIn | Both passed | DeepSeek V4.1 Flash, 2.2×took 2.9 s and 1.3 s | GPT-6 Luna, 3.3×cost $0.0001 and $0.0004 |
| Rewrite corporate jargon in plain words | Only GPT-6 Luna | DeepSeek V4.1 Flash, 2.7×took 2.4 s and 0.9 s | GPT-6 Luna, 2.9×cost $0.0001 and $0.0003 |
| Decline a meeting and offer two times | Both passed | GPT-6 Luna, 1.8×took 1.3 s and 2.2 s | GPT-6 Luna, 3.0×cost $0.0001 and $0.0002 |
| A product announcement with five rules | Both passed | DeepSeek V4.1 Flash, 1.8×took 1.5 s and 0.8 s | GPT-6 Luna, 3.7×cost $0.0001 and $0.0003 |
| Argue both sides of free buses | Both passed | Closetook 2.5 s and 2.6 s | GPT-6 Luna, 8.2×cost $0.0001 and $0.0011 |
| A discount, then sales tax | Both passed | DeepSeek V4.1 Flash, 4.4×took 3.1 s and 0.7 s | GPT-6 Luna, 1.8×cost $0.0001 and $0.0001 |
| Pens at 3 for $4 | Only DeepSeek V4.1 Flash | DeepSeek V4.1 Flash, 3.1×took 2.9 s and 0.9 s | GPT-6 Luna, 2.9×cost $0.0001 and $0.0004 |
| Compound interest over three years | Both passed | Closetook 2.0 s and 2.1 s | GPT-6 Luna, 2.4×cost $0.0001 and $0.0002 |
| Four-digit numbers whose digits sum to 9 | Both passed | DeepSeek V4.1 Flash, 2.9×took 2.0 s and 0.7 s | GPT-6 Luna, 2.8×cost $0.0001 and $0.0003 |
| The highest of three dice is a 5 | Both passed | DeepSeek V4.1 Flash, 2.1×took 2.2 s and 1.0 s | GPT-6 Luna, 2.4×cost $0.0001 and $0.0002 |
| An article in three bullets | Only DeepSeek V4.1 Flash | DeepSeek V4.1 Flash, 2.8×took 1.5 s and 0.5 s | GPT-6 Luna, 2.1×cost $0.0001 and $0.0002 |
| An email thread in one sentence | Both passed | DeepSeek V4.1 Flash, 2.5×took 1.1 s and 0.4 s | GPT-6 Luna, 1.9×cost $0.0001 and $0.0001 |
| Decisions and action items from a meeting | Both passed | GPT-6 Luna, 1.6×took 1.5 s and 2.4 s | GPT-6 Luna, 2.0×cost $0.0001 and $0.0002 |
| A quarterly memo for the CEO | Both passed | DeepSeek V4.1 Flash, 3.5×took 1.5 s and 0.4 s | GPT-6 Luna, 2.2×cost $0.0001 and $0.0003 |
| A study with a negative result | Both passed | DeepSeek V4.1 Flash, 2.6×took 1.3 s and 0.5 s | GPT-6 Luna, 2.6×cost $0.0001 and $0.0002 |
| The region with the most revenue | Both passed | DeepSeek V4.1 Flash, 3.2×took 2.5 s and 0.8 s | GPT-6 Luna, 3.9×cost $0.0001 and $0.0005 |
| Average order value in August | Both passed | DeepSeek V4.1 Flash, 3.1×took 2.6 s and 0.9 s | GPT-6 Luna, 3.4×cost $0.0001 and $0.0004 |
| Revenue change from July to August | Both passed | DeepSeek V4.1 Flash, 2.2×took 2.5 s and 1.1 s | GPT-6 Luna, 5.2×cost $0.0001 and $0.0007 |
| A median, filtered two ways | Both passed | DeepSeek V4.1 Flash, 4.0×took 2.5 s and 0.6 s | GPT-6 Luna, 2.7×cost $0.0001 and $0.0003 |
| Correlation between ad spend and sign-ups | Both passed | DeepSeek V4.1 Flash, 1.2×took 6.7 s and 5.8 s | GPT-6 Luna, 6.2×cost $0.0004 and $0.0023 |
| A late order | Both passed | DeepSeek V4.1 Flash, 1.2×took 1.6 s and 1.3 s | GPT-6 Luna, 5.4×cost $0.0001 and $0.0005 |
| A return inside the window | Both passed | DeepSeek V4.1 Flash, 1.4×took 1.2 s and 0.9 s | GPT-6 Luna, 3.2×cost $0.0001 and $0.0003 |
| A frustrated customer | Only DeepSeek V4.1 Flash | DeepSeek V4.1 Flash, 1.1×took 1.4 s and 1.3 s | GPT-6 Luna, 5.6×cost $0.0001 and $0.0005 |
| A refund request outside the window | Both passed | DeepSeek V4.1 Flash, 1.2×took 1.3 s and 1.1 s | GPT-6 Luna, 3.7×cost $0.0001 and $0.0003 |
| A message with a planted instruction | Only GPT-6 Luna | GPT-6 Luna, 1.5×took 1.5 s and 2.3 s | GPT-6 Luna, 8.5×cost $0.0001 and $0.0008 |
| A delivery message into Spanish | Both passed | DeepSeek V4.1 Flash, 2.8×took 2.6 s and 0.9 s | GPT-6 Luna, 3.5×cost $0.0001 and $0.0003 |
| A product description into French | Both passed | GPT-6 Luna, 1.5×took 1.4 s and 2.1 s | GPT-6 Luna, 9.9×cost $0.0001 and $0.0009 |
| A meeting note into German | Both passed | DeepSeek V4.1 Flash, 1.3×took 1.7 s and 1.3 s | GPT-6 Luna, 6.8×cost $0.0001 and $0.0005 |
| Idioms into natural Japanese | Both passed | GPT-6 Luna, 1.1×took 1.6 s and 1.7 s | GPT-6 Luna, 6.1×cost $0.0001 and $0.0006 |
| A lease clause into Brazilian Portuguese | Both passed | DeepSeek V4.1 Flash, 1.4×took 1.5 s and 1.1 s | GPT-6 Luna, 5.2×cost $0.0001 and $0.0005 |
| Customers in one country | Both passed | DeepSeek V4.1 Flash, 3.7×took 1.0 s and 0.3 s | GPT-6 Luna, 1.9×cost $0.0001 and $0.0001 |
| Count orders by status | Both passed | DeepSeek V4.1 Flash, 3.4×took 0.9 s and 0.3 s | GPT-6 Luna, 1.2×cost $0.0001 and $0.0001 |
| Revenue by category | Both passed | DeepSeek V4.1 Flash, 2.9×took 1.2 s and 0.4 s | GPT-6 Luna, 1.7×cost $0.0001 and $0.0002 |
| Every customer, even those without orders | Both passed | DeepSeek V4.1 Flash, 1.3×took 2.4 s and 1.9 s | GPT-6 Luna, 3.8×cost $0.0001 and $0.0004 |
| Monthly revenue with a running total | Both passed | DeepSeek V4.1 Flash, 2.5×took 1.8 s and 0.7 s | GPT-6 Luna, 2.2×cost $0.0001 and $0.0003 |
| A fact from one section | Both passed | DeepSeek V4.1 Flash, 4.3×took 1.3 s and 0.3 s | GPT-6 Luna, 2.3×cost $0.0001 and $0.0002 |
| Core hours and start times | Both passed | DeepSeek V4.1 Flash, 2.9×took 1.0 s and 0.3 s | GPT-6 Luna, 1.6×cost $0.0001 and $0.0001 |
| Two sections in one answer | Both passed | DeepSeek V4.1 Flash, 4.1×took 1.4 s and 0.3 s | GPT-6 Luna, 1.7×cost $0.0001 and $0.0001 |
| A later amendment changes the answer | Both passed | Closetook 1.0 s and 1.0 s | GPT-6 Luna, 3.4×cost $0.0001 and $0.0003 |
| A question the handbook doesn't answer | Both passed | DeepSeek V4.1 Flash, 4.7×took 1.9 s and 0.4 s | GPT-6 Luna, 1.8×cost $0.0001 and $0.0001 |
| Pick the tool and work out the date | Both passed | DeepSeek V4.1 Flash, 6.2×took 2.1 s and 0.3 s | GPT-6 Luna, 2.0×cost $0.0001 and $0.0001 |
| Convert a currency | Both passed | DeepSeek V4.1 Flash, 3.9×took 1.1 s and 0.3 s | GPT-6 Luna, 1.8×cost $0.0001 and $0.0001 |
| Book a meeting from a sentence | Both passed | DeepSeek V4.1 Flash, 1.9×took 1.2 s and 0.6 s | GPT-6 Luna, 3.5×cost $0.0001 and $0.0003 |
| Search, but don't book | Both passed | DeepSeek V4.1 Flash, 6.2×took 2.2 s and 0.4 s | GPT-6 Luna, 2.1×cost $0.0001 and $0.0002 |
| Two calls with a unit conversion | Both passed | DeepSeek V4.1 Flash, 3.0×took 1.9 s and 0.6 s | GPT-6 Luna, 2.9×cost $0.0001 and $0.0003 |
GPT-6 Luna vs DeepSeek V4.1 Flash in our test runs
GPT-6 Luna and DeepSeek V4.1 Flash on the same prompts, job by job: how many replies passed their check.
Based on 100 of our test runs on , through OpenRouter with the app's own prompt and settings. Updated .
| Job | GPT-6 Luna | DeepSeek V4.1 Flash |
|---|---|---|
| Coding | 5 of 5 | 5 of 5 |
| Writing | 5 of 5 | 4 of 5 |
| Math | 4 of 5 | 5 of 5 |
| Summarization | 4 of 5 | 5 of 5 |
| Data analysis | 5 of 5 | 5 of 5 |
| Customer support | 4 of 5 | 4 of 5 |
| Translation | 5 of 5 | 5 of 5 |
| SQL | 5 of 5 | 5 of 5 |
| RAG and answering from documents | 5 of 5 | 5 of 5 |
| Agents and tool use | 5 of 5 | 5 of 5 |
| All jobs | 47 of 50 · 1 message each · $0.0001 a reply | 48 of 50 · 1 message each · $0.0005 a reply |
The same prompts, sent to both through llmwise's own pipeline and checked the same way: five for each job. A bold count is the one that passed more. Messages: what each reply counted as on Pro. Cost: what OpenRouter charged us per reply, on average. Every prompt, and how it's scored.
Where they differed most, writing: both replies to a prompt only one of them passed.
An everyday prompt: “Rewrite corporate jargon in plain words”, in full.
GPT-6 Luna
Passed: Graded 4.7 of 5 on average (lowest 4).
In Q3, teams across the company worked together to focus more on customers. Retention measures improved by 12%, and the rate of customer loss fell by 0.3 percentage points in the enterprise segment. The company also adjusted its customer contact channels to make better use of available capacity.
417 tokens in, 124 out (59 of them reasoning) · 2.4 s · $0.0001 · 1 message on Pro · answered by openai/gpt-6-luna via OpenAI ·
DeepSeek V4.1 Flash
Failed: Graded 3.7 of 5 on average (lowest 3).
In the third quarter, teams from across the company worked together on a project to make the customer the main focus. The results: retention measures rose 12%, and the rate at which customers left fell by 30 basis points (0.30 percentage points) in the enterprise segment. We also adjusted the size of our channels — online, in-store, phone, and others — so we could use our resources more efficiently.
459 tokens in, 195 out (108 of them reasoning) · 0.9 s · $0.0003 · 1 message on Pro · answered by deepseek/deepseek-v4.1-flash via Together ·
GPT-6 Luna and DeepSeek V4.1 Flash on every plan
Whether the one-time free trial reaches each model, then each paid plan's messages on it.
| Plan | Price | GPT-6 Luna | DeepSeek V4.1 Flash |
|---|---|---|---|
| Free | $0 | Not in the trial | Not in the trial |
| Plus | $10 a month | 30 a day | 30 a day |
| Pro | $20 a month | 60 a day | 60 a day |
| Max | $50 a month | 120 a day | 120 a day |
| Ultra | $100 a month | 200 a day | 200 a day |
| Studio | $200 a month | 200 a day | 200 a day |
Prices don't include tax, which is added where it applies and shown before you pay. A paid plan's month is one allowance shared by every model, so each monthly count is the most you get if all of it goes to that model. It renews each billing period; everyday models refill daily at 00:00 UTC. Long chats count more per reply. How pricing works.
Every limit is published. Paid plans also have a monthly fair-use limit on AI cost: Plus $3.75, Pro $7.50, Max $20, Ultra $42, Studio $85. Using every message on your plan at typical sizes stays under it; very large messages and heavy research use it faster. Every limit, explained.
What differs
Messages on Pro
Both are everyday models: each message comes from Pro's 60 a day, so they cost the same.
Context window
Both take up to 1.05M tokens of context. A chat in llmwise holds up to 200k tokens, which fits in either, so the difference shows only through each maker's own API.
Images and PDFs
Both read images. DeepSeek V4.1 Flash gets a PDF's text rather than the file itself.
On Free
Neither is in the free trial: both need a paid plan, from Plus.
Where messages go
GPT-6 Luna: Sent to OpenAI directly. DeepSeek V4.1 Flash: Served through OpenRouter, only by hosts that don't store or train on prompts. The maker's own endpoint is never asked.
Fact by fact
| Fact | GPT-6 Luna | DeepSeek V4.1 Flash |
|---|---|---|
| Context window | 1.05M tokens | 1.05M tokens |
| Reads images | Yes | Yes |
| PDFs | Whole file | Text only |
| Reasoning | Yes | Yes |
| API price (October 2026) | $0.10 in / $0.50 out per million tokens | $0.30 in / $1.20 out per million tokens |
| A typical message at API prices (4,000 tokens in, 700 out) | $0.0008 | $0.0020 |
| A $10 top-up adds | Nothing: an everyday model's count is daily | Nothing: an everyday model's count is daily |
| Where a message goes | Sent to OpenAI directly. | Served through OpenRouter, only by hosts that don't store or train on prompts. The maker's own endpoint is never asked. |
| If the provider fails | If OpenAI fails before the reply starts (an overload, a server error, a dropped connection), llmwise sends the same request to GPT-6 Luna through OpenRouter instead. | When one host is down, OpenRouter moves the request to another host that meets the same rules. |
| Anthropic's safety fallback | Doesn't apply | Doesn't apply |
GPT-6 Luna or DeepSeek V4.1 Flash?
From the facts above and our test runs: the rest is how their answers suit your work, which one chat can show you.
Pick GPT-6 Luna: it reads a PDF as the whole file, charts and scans included; it costs its maker less to run ($0.0008 a typical message at API prices), though in llmwise the count is the same.
Each model's page, the families, and other pairs
GPT-6 Luna vs DeepSeek V4.1 Flash is one pair of models. The page below covers the whole families.
- GPT-6 Luna: price, limits and messages on every plan
- DeepSeek V4.1 Flash: price, limits and messages on every plan
- DeepSeek vs ChatGPT
- Claude Haiku 5.5 vs GPT-6 Luna
- GPT-6 Sol vs GPT-6 Luna
- Claude Haiku 4.5 vs GPT-6 Luna
- Claude Haiku 5.5 vs DeepSeek V4.1 Flash
- Gemini 3.8 Flash vs DeepSeek V4.1 Flash
- Claude Haiku 4.5 vs DeepSeek V4.1 Flash
- Every model-vs-model page
GPT-6 Luna is a GPT model; DeepSeek V4.1 Flash is a DeepSeek model.
Questions
Is GPT-6 Luna or DeepSeek V4.1 Flash cheaper in llmwise?
Both are everyday models: each message comes from Pro's 60 a day, so they cost the same. Every paid plan's monthly allowance is shared by the models on it, so each count is the most you get if it all goes to that model.
Can I try GPT-6 Luna and DeepSeek V4.1 Flash for free?
Neither is in the free trial of 5 messages: both need a paid plan, from Plus.
Which has the bigger context window, GPT-6 Luna or DeepSeek V4.1 Flash?
Neither: both take 1.05M tokens. A chat in llmwise holds up to 200k tokens, which fits in either, so the difference shows only through each maker's own API.
Can I use GPT-6 Luna and DeepSeek V4.1 Flash in the same chat?
Yes. Pick GPT-6 Luna for one message and DeepSeek V4.1 Flash for the next; the second sees the whole chat, including the first one's answer.
Which did better in your test runs, GPT-6 Luna or DeepSeek V4.1 Flash?
On the same 50 prompts, run on October 8, 2026, GPT-6 Luna passed 47 and DeepSeek V4.1 Flash passed 48. The table on this page has each job, and every reply is published.
Claude, GPT, Gemini, DeepSeek, Grok, Kimi, GLM, and Mistral, in one chat.
See what a message costs before you send it. Free is 5 messages to try; sign in with an email link, no password or card.