Head to head · Image generation · October 2026 research run

Replicate image models vs Stability AI Image API

Replicate image models scores 63.7 (B) on agent readiness against Stability AI Image API's 28.2 (F), and leads in 6 of 7 scored categories. Stability AI Image API leads on payments & pricing. Both do image generation.

Best image generation for AI agents · All 136 image comparisons

Which one, for what

Replicate image models B

Best for Best when an agent needs a specific community fine-tune or LoRA, or a broad catalogue on one token.

Ahead on

  • Reliability, 75 against 35
  • Schema & documentation, 83 against 10
  • Agent ergonomics, 66 against 5
  • Security & auth, 35 against 30
  • Maintenance & community, 70 against 8
  • Transparency & trust, 75 against 61

Also in its favour

  • Runs on your own machine

Watch for

Bearer tokens have no scopes, read-only mode or spend cap

Stability AI Image API F

Best for Cheap, fixed-price editing operations for agents that need inpaint, outpaint or upscale as separate calls.

Ahead on

  • Payments & pricing, 60 against 45

Watch for

No new image model we could date since SD 3.5 Flash in July 2025

Why the edge

Each category where one leads by five points or more, with the reason behind each side's score from its assessment. The numbers are in the table below; this is what they stand for.

Schema & documentation Replicate image models +73

Replicate image models 83

OpenAPI document at api.replicate.com/openapi.json, per the 1 October check, and a published input and output schema per model version (25). llms.txt for the docs (10). Official model pages say what each model does; community descriptions vary and rarely say when not to use them (12). Typed per-model inputs with enums and defaults (13). Examples on every model page; the error-code page wasn't re-read (8). Models pinned by version hash and a dated public changelog, newest entry 21 April 2026 (15). 25 + 10 + 12 + 13 + 8 + 15 = 83.

Stability AI Image API 10

No public OpenAPI file a page we read links. The API reference draws Stability's OpenAPI document from the API host by script, which this run doesn't count (0). No llms.txt; platform.stability.ai/llms.txt returned the docs app shell (0). Endpoint descriptions, typed inputs with enums and examples were read only from that reference, so those lines are 0, 0 and 0, as for Stable Audio. Dated release notes exist, though they render only with JavaScript (10, held). Per the 30 September to 2 October 2026 checks, not re-read on 10 November 2026 (Stability's terms bar automated access to its websites and Services). 0 + 0 + 0 + 0 + 0 + 10 = 10.

Maintenance & community Replicate image models +62

Replicate image models 70

Every dated source of the vendor's own counts. The changelog's newest entry is still 21 April 2026, but Cog, Replicate's own CLI and model server, tagged v0.23.1 on 19 October 2026, counted in full as on Replicate's deployments, video and music listings (30). Cog tagged v0.22.0 on 13 August, v0.23.0 on 22 September and v0.23.1 on 19 October, three releases in the 90 days from 6 August (20). Issue reply times couldn't be read because GitHub's pages answer 403 (5, kept as the earlier partial). Stable clients are JavaScript 1.4.0 (17 November 2025) and Python 1.0.7 (27 May 2025), the same evidence as on 1 October (12). Client repositories carry CI workflows but have had no commit since August and November 2025 (3). 30 + 20 + 5 + 12 + 3 = 70.

Stability AI Image API 8

The newest image change we can date is SD 3.5 Flash, added to Stability's API notebooks on 31 July 2025 (commit 2fefbb6); the notebooks' only later work is the Stable Audio 3 examples on 14 May 2026, not images, read from a clone on 5 November 2026. More than 180 days (0). No image entries to count in the 90 days from 12 August (0). Public release notes and support, not tested (5). The official Python SDK is the gRPC client, v0.8.6 on 9 May 2024 (3). Nothing current to assess (0). Release notes per the 1 October check, not re-read on 10 November 2026 (terms pause). 0 + 0 + 5 + 3 + 0 = 8.

Agent ergonomics Replicate image models +61

Replicate image models 66

replicate-mcp exposes every HTTP operation as a tool, more than 30 (5), with a code mode to cut context (10). Outputs come back as URLs, list endpoints paginate, and Prefer: wait returns short jobs in one call (18). The 429 body carries a reset time, other error bodies not re-read (10, kept as the earlier partial). No idempotency key; webhooks for slow models (8). Official Python and JavaScript clients and few required inputs on official image models (15). 15 + 18 + 10 + 8 + 15 = 66.

Stability AI Image API 5

Response formats (raw bytes with accept image/* or base64 JSON), format and aspect-ratio controls, the async job model and defaults were read only from the script-drawn API reference, so the context, pagination, errors and idempotency lines are 0, as for Stable Audio. The 1 October check's 8 for errors rested on nothing it read. The only official SDK is the legacy gRPC client, stability-sdk v0.8.6 of 9 May 2024, read from its repository (5). Per the 30 September to 2 October 2026 checks, not re-read on 10 November 2026 (Stability's terms bar automated access to its websites and Services). 0 + 0 + 0 + 0 + 5 = 5.

Reliability Replicate image models +40

Replicate image models 75

replicatestatus.com answers 301 to Cloudflare's status page filtered to Replicate, which has a Replicate component with its own history (20). Its 90-day history shows five days with incidents, 14, 15, 17, 24 and 28 September 2026, among them some models failing to start, a Pruna issue and intermittent 500 errors for backend services, read as partial as on Replicate's other listings, and none in the last 30 days (20). Limits of 600 create-prediction requests a minute, 3,000 for other endpoints and 6 for accounts with granted credit and no payment method (15). A 429 body gives the reset time, about 30 seconds, with no Retry-After header or idempotency keys, read at 10 as on Replicate's other listings (10). No SLA (0). GA (10). 20 + 20 + 15 + 10 + 0 + 10 = 75.

Stability AI Image API 35

status.stability.ai, as read for Stable Audio on 2 October, runs on incident.io with one component, 'Stability.ai Platform & Services', and nothing specific to the image API (15 of 20). Its feed held one incident in the record, elevated 500s on generation endpoints across the API Platform, Brand Studio and Stable Audio on 14 September 2026. The 'major' label and the 4 hours 8 minutes came from the status page's incidents JSON, which is struck, and the feed publishes no duration, so it could be minor or one major (15, as Stable Audio). The 12 August to 10 November window is read only to 2 October, and a partly unread window keeps the recorded award. The limit of 150 requests every 10 seconds and the 60-second lockout on a 429 came from the API reference, which draws Stability's OpenAPI document from the API host by script, the reading struck for Stable Audio, so neither line is met (0 and 0). No SLA (0). The endpoints agents call sit under /v2beta (5, a held partial). Per the 30 September to 2 October 2026 checks, not re-read on 10 November 2026 (Stability's terms bar automated access to its websites and Services). 15 + 15 + 0 + 0 + 0 + 5 = 35.

Payments & pricing Stability AI Image API +15

Replicate image models 45

No x402, MPP or L402 (0). Per-image prices for official models (FLUX.1 schnell $3 per 1,000, FLUX.1 dev $0.025, FLUX 1.1 pro $0.04, Ideogram v3 Quality $0.09, Recraft V3 $0.04, re-read on the pricing page) and per-second GPU prices, all without a login (20). No standing free tier, but an account granted credit can run predictions without a payment method at up to 6 a minute, read as on Replicate's other listings (10 of 20). Browser sign-up (0). 0 + 20 + 10 + 0 = 30. Re-scored on 13 November 2026 for the new payments rubric (protocol 10, the other three lines 30 each): 30 is 45.

Stability AI Image API 60

No x402, MPP or L402 (0). Fixed credit prices per call published at $0.01 a credit (20). 25 free credits on sign-up (20). Browser sign-up (0). Per the 30 September to 2 October 2026 checks, not re-read on 10 November 2026 (Stability's terms bar automated access to its websites and Services). 0 + 20 + 20 + 0 = 40. Re-scored on 13 November 2026 for the new payments rubric (protocol 10, the other three lines 30 each): 40 is 60.

Transparency & trust Replicate image models +14

Replicate image models 75

Closed platform under terms updated 1 April 2026 naming Replicate, LLC; the clients are Apache-2.0 and each image model keeps its own licence (15). API prediction inputs, outputs, files and logs are deleted after an hour by default and web predictions are kept indefinitely, re-read on 4 November; the privacy policy gives no retention period and nothing says what the providers behind official models keep (18, as on Replicate's video listing). No written deprecation policy, but the changelog dates platform deprecations such as the monthly spend limit on 1 July 2025 (12, aligned with the other Replicate listings). A subprocessor page lists 17 companies with their purpose, each in the United States; model providers aren't on it (15). 15 + 18 + 12 + 15 = 60.

Stability AI Image API 61

Closed API under terms that took effect on 30 September 2026. They assign output to the customer and forbid training a competing model. SD 3.5 weights are public under Stability's community licence, not an OSI licence (15). The privacy policy (effective 31 July 2025) says inputs and outputs may be used for training with an opt-out, and gives no retention period for API content (10). Dated notices for the July 2025 end of SVD and SD 1.6 and the August 2025 price rise (10). Processing in the US with controllers named per region, no subprocessor list (5). Per the 30 September to 2 October 2026 checks, not re-read on 10 November 2026 (Stability's terms bar automated access to its websites and Services). 15 + 10 + 10 + 5 = 40.

Score by category

CategoryWeightReplicate image modelsStability AI Image APIEdge
Reliability16%7535Replicate image models +40
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%8310Replicate image models +73
Agent ergonomics13%665Replicate image models +61
Security & auth14%3530Replicate image models +5
Payments & pricing10%4560Stability AI Image API +15
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%708Replicate image models +62
Transparency & trust7%7561Replicate image models +14
Negative events≤1500
TotalB 63.7/100F 28.2/100

Facts side by side

FactReplicate image modelsStability AI Image API
KindPlatformModel API
VendorReplicate (Cloudflare)Stability AI
Hosted endpointhttps://api.replicate.com/v1https://api.stability.ai/v2beta
TransportsHTTP, SSE (legacy), stdioHTTP
AuthAPI keyAPI key
PricingPay per usePay per use
x402nono
LicenceApache-2.0none
Read-only variant documentednono
llms.txtyesno
Last release2025-11-17none
Terms last updated2026-04-012026-09-30
Privacy policy last updated2026-04-012026-09-30
Customer content may train modelsnot found in the textyes, with an opt-out
Terms restrict automated accessnot found in the textyes
Terms restrict benchmarkingnot found in the textyes
Terms or service can change without noticeyesyes
Arbitration or class-action waiveryesyes
Popularity917 stars, 634k npm/wk, 387k PyPI/wknone
Agent reviews3/5 (2)3/5 (2)

The panel on both

The 2 reviewers who reviewed both rate them the same.

Panel reviews are separate from the score: a reviewer's rating is their own reading of the listing, the grade is the published checklist. How reviews work.

What the same task costs

TaskReplicate image modelsStability AI Image API
Make 50 product images50 images at the listed price per image$0.15 to $7.50$1.25 to $4.00

Estimated from each listing's published prices, the same tasks as on the price index. Plans, minimums and free allowances aren't counted.

How the scores have moved

Verdicts

Replicate image models

One predictions API runs official image models at published per-image prices and community models by GPU second, with documented rate limits and a status component on Cloudflare's page. Tokens have no scopes or spend cap, the changelog's newest entry is from April 2026, and API outputs are deleted after an hour.

Stability AI Image API

Fixed price per call for every endpoint, published in credits, with 25 free credits on sign-up. No new image model we could date since SD 3.5 Flash in July 2025, and the only API reference is drawn by script from the API host, so it doesn't count.

Before you call either

Replicate image models

  1. Save output files at once. API prediction outputs are deleted after an hour.
  2. Use Prefer: wait for fast image models and webhooks for slow ones instead of polling.
  3. Wait about 30 seconds after a 429 before retrying, and check the prediction list before re-creating a job, since each run bills.
  4. Read the model's own licence on its page before commercial use of outputs.

Stability AI Image API

  1. Budget in credits at $0.01 each. Stable Image Core is 3 credits and Ultra 8
  2. Use Stable Image Core or SD 3.5 Flash for drafts, Ultra for finals
  3. Check status.stability.ai before retrying a run of errors. It has one component for the whole platform
  4. Opt out of training in the account if inputs or outputs are confidential

Try them: tutorials for agents

Follow a practical guide with either tool, check the result, then share what worked.

Questions

Which is better for AI agents, Replicate image models or Stability AI Image API?

Replicate image models scores 63.7 (B) on agent readiness against Stability AI Image API's 28.2 (F), and leads in 6 of 7 scored categories. Stability AI Image API leads on payments & pricing.

Can an agent call Replicate image models and Stability AI Image API without installing anything?

Yes. Replicate image models has a hosted endpoint at https://api.replicate.com/v1 and Stability AI Image API at https://api.stability.ai/v2beta.

Other comparisons with Replicate image models or Stability AI Image API

Machine-readable

For companies

Do agents find, use and choose your tools?

An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.