Y-API vs DeepInfra
deepseek/deepseek-v4-flash, minimax/minimax-m2.7, xiaomi/mimo-v2.5 — Free — no credit deducted. Every call to these models deducts nothing from your balance. Same base_url, same key, same SDK — the model field is the only thing you change, so they are also the fastest way to try Y-API before you top up. DeepInfra is another gateway reselling the same vendors’ models, so this is not two labs competing — it is one question. For the 15 models both of us sell, does going through Y-API cost less? Both columns below are cash per 1M input tokens, and DeepInfra’s side is the cheapest tier we could find on its own price list, which makes every multiple a floor rather than a headline.
| Dimension | Y-API | DeepInfra |
|---|---|---|
| API protocol | Y-APIOpenAI Chat Completions. Point base_url at https://api.y-api.bestvirtualgoods.com/v1, swap the key, leave the rest of your code alone. | DeepInfraAn OpenAI-compatible endpoint of its own, on a different base URL and with different model strings — see the row for each model below. |
| Models on one key | Y-API47 model IDs from 10 vendors, one balance, one key. | DeepInfraIts own published catalog. Of the 47 models we compare against it, 15 are priced there. |
| deepseek/deepseek-v4-flash — Input / 1M tokens | Y-API$0.00 — Free — no credit deducted | DeepInfra$0.09 (deepseek-ai/DeepSeek-V4-Flash) · standard tier; priority and flex tiers differ |
| minimax/minimax-m2.7 — Input / 1M tokens | Y-API$0.00 — Free — no credit deducted | DeepInfra$0.25 (MiniMaxAI/MiniMax-M2.7) |
| xiaomi/mimo-v2.5 — Input / 1M tokens | Y-API$0.00 — Free — no credit deducted | DeepInfra$0.14 (XiaomiMiMo/MiMo-V2.5) |
| deepseek/deepseek-v4-pro — input / 1M tokens | Y-API$0.05 in cash | DeepInfra$1.30 as deepseek-ai/DeepSeek-V4-Pro — 26× ours (standard tier; priority and flex tiers differ) |
| anthropic/claude-opus-4.7 — input / 1M tokens | Y-API$0.50 in cash | DeepInfra$5.00 as anthropic/claude-opus-4-7 — 10× ours |
| anthropic/claude-haiku-4.5 — input / 1M tokens | Y-API$0.10 in cash | DeepInfra$1.00 as anthropic/claude-haiku-4-5 — 10× ours |
| google/gemini-3.1-pro-preview — input / 1M tokens | Y-API$0.20 in cash | DeepInfra$2.00 as google/gemini-3.1-pro — 10× ours (listed under a different name) |
| google/gemini-3.1-flash-lite — input / 1M tokens | Y-API$0.025 in cash | DeepInfra$0.25 as google/gemini-3.1-flash-lite — 10× ours |
| minimax/minimax-m3 — input / 1M tokens | Y-API$0.03 in cash | DeepInfra$0.28 as MiniMaxAI/MiniMax-M3 — 9.33× ours |
| xiaomi/mimo-v2.6-pro — input / 1M tokens | Y-API$0.05 in cash | DeepInfra$0.43 as XiaomiMiMo/MiMo-V2.6-Pro — 8.6× ours |
| qwen/qwen3.8-max — input / 1M tokens | Y-API$0.20 in cash | DeepInfra$1.65 as Qwen/Qwen3.8-Max — 8.25× ours (standard tier; priority and flex tiers differ) |
| z-ai/glm-5.1 — input / 1M tokens | Y-API$0.14 in cash | DeepInfra$1.05 as zai-org/GLM-5.1 — 7.5× ours |
| google/gemini-3.5-flash — input / 1M tokens | Y-API$0.20 in cash | DeepInfra$1.50 as google/gemini-3.5-flash — 7.5× ours |
| qwen/qwen3.8-flash — input / 1M tokens | Y-API$0.02 in cash | DeepInfra$0.113 as Qwen/Qwen3.8-Flash — 5.65× ours |
| google/gemini-3.7-flash — input / 1M tokens | Y-API$0.15 in cash | DeepInfra$0.75 as google/gemini-3.7-flash — 5× ours |
| What a top-up converts to | Y-API$1 becomes $10 of account credit, then credit is deducted at each model’s own rate. | DeepInfraYou pay the list price in cash. There is no credit multiplier, and nothing to expire. |
| How the rate is structured | Y-APIOne rate per model — every hour, every region, every request. | DeepInfraThe figures above sit inside tiers: standard tier; priority and flex tiers differ. We quoted the cheapest of them. |
| Cache-hit pricing | Y-APINo cache discount. A repeated prompt is billed exactly like a new one. | DeepInfraNo cached-input rate published for the rows above, so both columns assume every request is a cache miss. If DeepInfra discounts cache hits off the rate card, the gap narrows. |
| Models it does not price | Y-APIAll 47 are callable on the same key. | DeepInfraWe looked for 15 models and could not find them (minimax/minimax-m3.1-flash-preview, anthropic/claude-opus-4.6, anthropic/claude-opus-5.5, anthropic/claude-haiku-5.5, google/gemini-4-argon, google/gemini-3.8-flash, openai/gpt-5.5-pro, openai/gpt-5.5, openai/gpt-5.4, openai/gpt-6-sol, openai/gpt-6.1-sol, openai/gpt-5.4-mini, openai/gpt-6-luna, x-ai/grok-4.6 and x-ai/grok-4.7); we have not checked 17 models against it yet. |
Against us
Where DeepInfra costs less
A comparison that only runs one way is an advertisement. Everything below comes from the same price list as the figures above. Free status is set by our upstream provider and can be withdrawn — when it is, these calls start billing at the listed rate and your code does not change. The models page always shows the current state.
We have no cache discount at all
For paid models, every request is billed at the full rate whether or not the prompt is identical to the last one. Long fixed system prompts and repeated document context are exactly where that hurts.
Some rows are not the same build on both sides
We price google/gemini-3.1-pro-preview against its google/gemini-3.1-pro. Matching those by vendor, parameter count and release is our judgement, not their claim. The prices are comparable; the behaviour may not be. Pin the model ID and test it if that matters to you.
DeepInfra sells far more models than we do
Its catalog spans vendors and open-weight models we do not carry at all, and it can route around a provider that is down. Our 47 are a deliberate shortlist; if the model you need is not on it, our price for the others is irrelevant to you.
Fit
Which one to call
Call it through Y-API if
- deepseek/deepseek-v4-flash, minimax/minimax-m2.7, xiaomi/mimo-v2.5 — Free — no credit deducted
- Your input is mostly fresh, where the gap is 26× on deepseek/deepseek-v4-pro, 10× on anthropic/claude-opus-4.7, 10× on anthropic/claude-haiku-4.5, 10× on google/gemini-3.1-pro-preview, 10× on google/gemini-3.1-flash-lite, 9.33× on minimax/minimax-m3, 8.6× on xiaomi/mimo-v2.6-pro, 8.25× on qwen/qwen3.8-max, 7.5× on z-ai/glm-5.1, 7.5× on google/gemini-3.5-flash, 5.65× on qwen/qwen3.8-flash and 5× on google/gemini-3.7-flash at the current top-up rate.
- You want one balance and one key across 47 models from 10 vendors instead of an account per vendor.
- You would rather not track which tier, region or hour you are being billed under. We have one rate per model.
Go to DeepInfra if
- You need a model outside our 47, or you want automatic fallback across providers when one is down.
- You want the billing relationship, the SLA and the support to be with DeepInfra rather than with a reseller.
FAQ
Questions people ask before switching
- Free models
- deepseek/deepseek-v4-flash, minimax/minimax-m2.7, xiaomi/mimo-v2.5 — Free — no credit deducted. Every call to these models deducts nothing from your balance. Same base_url, same key, same SDK — the model field is the only thing you change, so they are also the fastest way to try Y-API before you top up. Free status is set by our upstream provider and can be withdrawn — when it is, these calls start billing at the listed rate and your code does not change. The models page always shows the current state.
- Is Y-API cheaper than DeepInfra?
- On fresh input, yes: deepseek/deepseek-v4-pro runs $0.05 against their $1.30, anthropic/claude-opus-4.7 runs $0.50 against their $5.00, anthropic/claude-haiku-4.5 runs $0.10 against their $1.00, google/gemini-3.1-pro-preview runs $0.20 against their $2.00, google/gemini-3.1-flash-lite runs $0.025 against their $0.25, minimax/minimax-m3 runs $0.03 against their $0.28, xiaomi/mimo-v2.6-pro runs $0.05 against their $0.43, qwen/qwen3.8-max runs $0.20 against their $1.65, z-ai/glm-5.1 runs $0.14 against their $1.05, google/gemini-3.5-flash runs $0.20 against their $1.50, qwen/qwen3.8-flash runs $0.02 against their $0.113 and google/gemini-3.7-flash runs $0.15 against their $0.75. Two things bound that. Their side is always the cheapest tier they publish, so each multiple is a floor rather than a peak; and the gap depends on the 1:10 top-up rate. For scale: our cheapest paid model starts at $0.01 per 1M input tokens.
- Why is it cheaper — are you buying below their price?
- No, and the honest answer matters here. For paid models, our catalog rates, read as credit, sit close to DeepInfra’s cash list prices. Almost the whole gap is the top-up conversion: $1 buys $10 of credit.
- Which models does DeepInfra price against ours?
- It prices 15 of our 47, with the name each one carries there: deepseek/deepseek-v4-flash (deepseek-ai/DeepSeek-V4-Flash there), minimax/minimax-m2.7 (MiniMaxAI/MiniMax-M2.7 there), xiaomi/mimo-v2.5 (XiaomiMiMo/MiMo-V2.5 there), deepseek/deepseek-v4-pro (deepseek-ai/DeepSeek-V4-Pro there), anthropic/claude-opus-4.7 (anthropic/claude-opus-4-7 there), anthropic/claude-haiku-4.5 (anthropic/claude-haiku-4-5 there), google/gemini-3.1-pro-preview (google/gemini-3.1-pro there), google/gemini-3.1-flash-lite (google/gemini-3.1-flash-lite there), minimax/minimax-m3 (MiniMaxAI/MiniMax-M3 there), xiaomi/mimo-v2.6-pro (XiaomiMiMo/MiMo-V2.6-Pro there), qwen/qwen3.8-max (Qwen/Qwen3.8-Max there), z-ai/glm-5.1 (zai-org/GLM-5.1 there), google/gemini-3.5-flash (google/gemini-3.5-flash there), qwen/qwen3.8-flash (Qwen/Qwen3.8-Flash there) and google/gemini-3.7-flash (google/gemini-3.7-flash there). Those are the only rows in the table above — a row with a published price on one side only is not a comparison.
- What about the models it does not price?
- We looked for 15 models and could not find them (minimax/minimax-m3.1-flash-preview, anthropic/claude-opus-4.6, anthropic/claude-opus-5.5, anthropic/claude-haiku-5.5, google/gemini-4-argon, google/gemini-3.8-flash, openai/gpt-5.5-pro, openai/gpt-5.5, openai/gpt-5.4, openai/gpt-6-sol, openai/gpt-6.1-sol, openai/gpt-5.4-mini, openai/gpt-6-luna, x-ai/grok-4.6 and x-ai/grok-4.7); we have not checked 17 models against it yet. We keep “we did not look” and “they do not sell it” apart on purpose: they are different claims, and only the first one is ours to make.
- Is the model on their side literally the same build?
- Not for google/gemini-3.1-pro-preview (google/gemini-3.1-pro there) — a dated snapshot or a differently named build, matched by vendor, parameter count and release date. Compare the prices; do not assume identical behaviour.
- How much of my code has to change?
- base_url, the key, and the model string. Both sides speak OpenAI Chat Completions, so nothing else in the request body changes — ours are prefixed, for example deepseek/deepseek-v4-flash.
- Is Y-API up right now — where is the status page?
- Availability is not a row in the table above because it changes by the hour. Live status of the website, the gateway and account sign-up, plus incident history and 7/30-day uptime, is on the status page at https://y-api.bestvirtualgoods.com/status.
Provenance
Sources
Every DeepInfra figure on this page was read from the page below on 2026-10-11, at the cheapest tier it publishes. Our own figures are on our pricing and terms pages.
Point one call at it
Change base_url, the key, and the model string on a single call you already make. The request log will tell you what it costs before you move anything else.