Blog cover generator.
Every post gets art.
The founding Pixfaro use case: your agent drafts the post, calls one endpoint for the cover, and publishes — every major image model, one prepaid balance.
Works with Claude Desktop · Claude Code · Cursor · Windsurf · any OpenAI SDK
Real prompts. Real outputs.
Nothing cherry-picked twice.
Every image is a real Pixfaro output for the exact prompt beneath it — generated through the API itself, model named in the caption, click for full resolution. Regenerated when models update.
An all-round blog-art workhorse, budget models for daily cadence, and typography-strong finals for covers that carry the title — one request shape across all of them.
One key. One balance.
Zero provider accounts.
A blog that publishes weekly needs art weekly, forever. At $0.004-class prices per image the cover stops being a task and becomes a line in the pipeline — and switching the model string upgrades the whole blog's art the week a better model lands.
This is the pipeline Pixfaro was built for: the agent drafts your post, calls the MCP for a cover that matches the argument, and publishes — one pipeline, zero tabs. Works in Claude Desktop, Claude Code, Cursor, or headless via the CLI in CI.
"args": ["-y", "@pixfaro/mcp"],
"env": { "PIXFARO_KEY": "pf_live_…" } }
# then: “add a cover image to this post, house style, wide”
Pin a style clause — “editorial, muted palette, single strong metaphor” — and reuse it in every prompt; the archive page starts looking art-directed instead of assembled. Want your handle on every cover? The overlay parameter composites your logo or @name pixel-exact, no model roulette.
Which model wins for blog covers?
Weekly benchmarks, real prices — the same registry the API bills from. Our per-model read for blog-cover work:
| Model | Our verdict | Median speed | Price / image | |
|---|---|---|---|---|
| nano-banana-2 | the blog-art workhorse — strong editorial metaphors per dollar | 10.7s | $0.080 | Try |
| flux-dev | richer texture for flagship posts, still far from frontier cost | 4.2s | $0.030 | Try |
| gemini-pro-image | when the title sits on the cover — real type, native wide | 20.8s | $0.164 | Try |
| flux-2-klein | daily-cadence covers at budget price, current-gen look | 0.8s | $0.006 | Try |
| flux-schnell | mood-boarding a series style for pennies | 2.4s | $0.004 | Try |
Speed and price come from our weekly benchmarks — provider cost + 20%, updated when providers move. Every model and tier → /pricing
Three steps.
About a minute.
Sign up for the $1 credit (no card) — or add the MCP to the agent that already writes your posts.
export PIXFARO_KEY=pf_live_…
Feed the post's central metaphor, not its title — covers that show the argument outperform covers that decorate it.
-H "Authorization: Bearer $PIXFARO_KEY" \
-d '{"model":"nano-banana-2","prompt":"blog cover, data
streams converging into a lighthouse beam, wide"}'
Add your logo or handle with the overlay parameter — composited pixel-exact by Pixfaro, never gambled on the model — and drop the returned URL into your front matter.
"position": "bottom-right" }
Pay per image.
Know the cost before you run.
A year of weekly covers — 52 posts, each with a few drafts and a keeper — costs less than one stock-photo subscription month. Per image it's $0.004 to $0.164 depending on the model, printed next to each one before you run.
Top up a balance, and each generation debits exactly the listed price — cost and latency come back in the response, so an agent (or a budget) always knows where it stands. Failed generations aren't charged.
Signup gives you $1 of credit with no card — enough to draft dozens of images on the budget models and still compare finals on a frontier one before you spend a cent.
Credits last 365 days; there's nothing to cancel. Optional auto-reload (with a monthly cap) keeps long agent runs from stalling mid-pipeline.
Fair questions,
straight answers.
Something else? support@pixfaro.com — a human answers.
What size should a blog cover / OG image be?
The de-facto OG standard is 1200×630. Generate a native wide frame on a Gemini-family model (or at 2K for headroom) and crop to 1.91:1 — the wide composition survives the crop; a square never does.
Can my agent really do this end to end?
Yes — it's the founding use case. Over MCP, “add a cover to this post” resolves to a generate call with your house style, and the response is a URL your static-site front matter can use directly. This very pattern runs Pixfaro's own blog art.
How do I keep covers consistent across a series?
Pin one style clause and append it to every cover prompt — same palette, same tone, different metaphor per post. Keep it in your repo next to the blog config so the agent (and the next human) uses it verbatim.
Can I put my blog's name or logo on every cover?
Yes, and not by hoping the model spells it right: the overlay parameter composites your uploaded logo or handle text into a corner, pixel-exact with automatic contrast. Upload once via POST /v1/logos, reference it in every request.
Are the images fine for commercial blogs and ads?
Yes, within the Terms and provider policies — you own your outputs as far as the law allows. Covers are also served from stable URLs, so you can hotlink or mirror to your own CDN, whichever your stack prefers.
What does a cover habit actually cost?
Three drafts plus a keeper on nano-banana-2 is about $0.32 per post; on the budget tier the same routine is under 3 cents. The exact debit rides back in every response, so the finance question has a receipt.
One balance.
Every kind of image.
The same wide-format discipline, pointed at videos instead of posts.
Wordmarks, monograms, mascots — the shoot-out approach to brand marks.
Die-cut characters and pack-ready art — same balance, same key.
Every model and resolution tier, priced per image — no subscription.
Building instead of browsing? The same models sit behind the REST API and the MCP server.