Head to head · Updated 9 Oct 2026
SayGM vs OpenRouter
OpenRouter passes list price through and takes 5.5 percent when you add credit. SayGM takes nothing on deposit and bills below list. On Claude, GPT and Gemini that makes SayGM 16 to 34 percent cheaper. OpenRouter keeps the larger catalogue and a free tier.
| OpenRouter | $105.50 |
|---|---|
| SayGM · Claude | $88.90 |
| SayGM · GPT | $74.50 |
| SayGM · Gemini | $69.50 |
Real-time calls to frontier models, OpenRouter credit bought by card.
Thirteen points, side by side.
Grouped by what usually drives the decision. A dot marks the side that comes out ahead; rows where both are the same stay in.
| Feature | SayGM | OpenRouter |
|---|---|---|
| Cost | ||
| Token price | 11–30.5% below maker list price (ahead) | Maker list price |
| Credit top-up fee | None (ahead) | 5.5% by card ($0.80 minimum), 5% by crypto |
| Subscription | None, pay as you go | None, pay as you go |
| Free usage | None | Free tier with tight limits (ahead) |
| Catalogue | ||
| Models | About 65, curated | Several hundred, including long-tail (ahead) |
| Confidential open models | Sealed in a TEE end to end (-tee IDs) | Via third-party TEE hosts such as Phala |
| API | ||
| API surface | Each model's native API | One OpenAI-compatible API |
| Prompt caching | Conversation kept on one provider | Depends on provider routing |
| Fallback | Retries plus cascade models | Automatic retry, provider ordering |
| Multi-model calls | Fusion models | Not offered |
| Rate limits | Per key, raised on request | By tier, tightest on free |
| Privacy | ||
| Gateway privacy | Intel TDX enclave, public attestation (ahead) | Privacy policy, opt-in logging |
| Frontier prompts | Reach the lab as with a direct call | Reach the lab as with a direct call |
01 · Price
16 to 34 percent less on every frontier model.
OpenRouter does not mark up tokens; it charges the maker's list price. SayGM charges below it: 11.1 percent under Anthropic, 25.5 percent under OpenAI and 30.5 percent under Google. Add OpenRouter's top-up fee and the gap widens to the figures above.
Where it does not hold. On open-weight models the comparison flips for input tokens: OpenRouter's cheapest providers for Kimi K3 and DeepSeek V4.1 Flash cost less, on hosts that are not confidential.
Worked example
A production app on Claude Sonnet 5.5 sends 50M input and 15M output tokens a month, uncached.
That costs $222 a month on SayGM and $264 on OpenRouter: $41 a month, or $497 a year.
Caching only pays out if a conversation stays on one provider. A router that spreads turns across providers loses the cache discount without reporting it.
- SayGM
- OpenRouter, fee included
Show data as a table
| Model | SayGM | OpenRouter, fee included | Difference |
|---|---|---|---|
| Claude Fable 5.1 | $16.00 | $18.99 | -16% |
| GPT-5.5 | $7.45 | $10.55 | -29% |
| Claude Opus 5 | $8.01 | $9.50 | -16% |
| Claude Opus 5.5 | $6.40 | $7.60 | -16% |
| GPT-5.6 Terra | $2.98 | $4.22 | -29% |
| Gemini 3.1 Pro Preview | $2.78 | $4.22 | -34% |
| Claude Sonnet 5.5 | $3.20 | $3.80 | -16% |
| Claude Sonnet 5 | $3.20 | $3.80 | -16% |
| GPT-6.1 Sol | $2.68 | $3.80 | -29% |
| Gemini 3.5 Flash | $2.08 | $3.16 | -34% |
| Kimi K3 | $1.76 | $2.24 | -21% |
| Claude Haiku 4.5 | $1.60 | $1.90 | -16% |
| GPT-5.6 Luna | $0.298 | $0.422 | -29% |
| GPT-6 Luna | $0.13 | $0.194 | -33% |
| DeepSeek V4.1 Flash | $0.288 | $0.13 | 122% |
Price your own workload
Uncached snapshot prices on every route.
Monthly bill by provider
- OpenRouter, fee included$211
- Requesty, 5% included$210
- Vercel AI Gateway$200
- Anthropic direct, list price$200
- SayGM$178
- A year, vs the cheapest alternative
- $265
- 11% lower than Vercel AI Gateway
- A year, vs OpenRouter
- $397
- Fee included
Uncached prices from the October 2026 snapshot. Prompt caching lowers every provider's bill; batch jobs bought direct cost half of list. Current SayGM rates are on saygm.com (opens in a new tab).
02 · Fees
OpenRouter's fee lands before a single token is served.
OpenRouter charges 5.5 percent on each card top-up, with a $0.80 minimum. The cost scales with how much credit is loaded, not with which model runs, so it is easy to miss on an invoice.
SayGM credits the full top-up to the balance, then bills tokens below list.
Where it does not hold. Crypto top-ups on OpenRouter cost 5 percent rather than 5.5.
Show data as a table
| Annual fee | |
|---|---|
| $50 a month | $33 |
| $200 a month | $132 |
| $500 a month | $330 |
| $2,000 a month | $1,320 |
| $5,000 a month | $3,300 |
03 · Privacy
Attestation on one side, a written policy on the other.
OpenRouter's privacy rests on a commitment the company honours. SayGM's rests on hardware isolation that anyone can verify, and it covers the routing layer as well as the model.
Where it does not hold. In both cases Claude, GPT and Gemini receive the prompt as they would from a direct API call.
SayGM
- Gateway runs inside an Intel TDX trusted execution environment.
- HTTPS terminates inside the enclave; operators and host sit outside it.
- A hardware-signed quote, rooted in Intel, can be checked against your own nonce.
- Logs hold metadata only, never prompts or completions.
- Confidential open-weight models are decrypted only inside the enclave.
OpenRouter
- Prompts processed under a published privacy policy.
- Logging is opt-in.
- No hardware guarantee for the gateway itself.
- Open models can reach TEE hosts through third parties.
04 · Trade-offs
Where each one is the better choice.
If most spend goes to Claude, GPT or Gemini, SayGM costs less and removes the fee. If the work depends on long-tail models or a free tier, OpenRouter fits better. Many teams run both.
SayGM is better for
- 01
Below list on frontier models
Claude, GPT and Gemini cost 16 to 34 percent less than OpenRouter's fee-inclusive rate.
- 02
No fee on credit
Top-ups are credited in full. On OpenRouter, 5.5% of each card deposit goes to the fee.
- 03
Native provider APIs
Anthropic and Gemini requests keep provider features such as prompt caching and extended thinking.
- 04
A gateway you can verify
The gateway runs in an Intel TDX enclave and publishes attestation, so its isolation can be checked rather than assumed.
OpenRouter is better for
- 01
Long-tail models
Several hundred models against SayGM's 65. Legacy and niche models are far more likely to be listed.
- 02
Free models
A free tier with tight limits, useful for prototypes. SayGM has none.
- 03
One request format
Every model speaks the OpenAI format, so code that hops between models needs no provider-specific handling.
- 04
Cheapest open-weight providers
OpenRouter's cheapest providers for Kimi K3 and DeepSeek V4.1 Flash cost less per input token. Those providers are not confidential.
05 · Switching
The move is a base URL and a key.
Leaving OpenRouter? SayGM is currently paying off the last OpenRouter bill of teams that switch. The terms are SayGM's own, so confirm them on its site before you count on it.
- 01
Create a key and load credit
The full top-up becomes balance. There is no subscription or minimum.
- 02
Change the base URL
Point your Anthropic, OpenAI or Gemini SDK at SayGM and use model IDs from its catalogue.
- 03
Keep a second provider
Leave your current gateway in place as a fallback for models SayGM does not list.
06 · Questions
SayGM vs OpenRouter questions.
What is SayGM?
SayGM is an LLM gateway. One API key reaches Claude, GPT, Gemini and a set of open-weight models. There is no subscription and no fee on credit top-ups. Prices are capped at the model maker's list price, and the gateway runs inside an Intel TDX trusted execution environment that publishes remote attestation.
Is SayGM cheaper than OpenRouter?
On the frontier models in this index, yes, by 16 to 34 percent once OpenRouter's 5.5 percent top-up fee is included. The exception is open-weight models: the cheapest OpenRouter providers for Kimi K3 and DeepSeek V4.1 Flash cost less on input tokens. Check the current rate for the model you use.
How does SayGM compare to Vercel AI Gateway, Requesty or buying direct?
Vercel AI Gateway and the labs themselves charge list price for real-time calls; Requesty adds 5 percent unless you bring your own keys. SayGM bills 11 to 30.5 percent below list depending on the lab. Buying direct still wins for batch jobs, which cost half of list, and for contracts such as HIPAA BAAs in your own name.
Does SayGM work with Cursor, Cline and Claude Code?
Yes. SayGM exposes OpenAI-compatible, Anthropic-compatible and Gemini-compatible endpoints. Any tool that accepts a custom base URL works after changing the URL and the API key.
How many models does SayGM support?
About 65 at the time of writing: the current Claude, GPT and Gemini lines plus open-weight models such as Kimi K3, GLM 5.3 and DeepSeek V4.1. OpenRouter lists several hundred. If you depend on a niche or legacy model, confirm it is in the SayGM catalogue first.
Can SayGM see my prompts?
The gateway runs in a trusted execution environment sealed from operators and host machines. Confidential open-weight models (IDs ending in -tee) also run inside the enclave, so prompts are decrypted only there. Frontier providers still receive the prompt, as with a direct API call. SayGM strips account identity and can replace personal data before sending.
Is there a free tier or minimum spend?
No free tier. SayGM is pay as you go: load credits, spend them per token. There is no fee on the top-up itself.
What is Bittensor subnet 28?
SayGM runs on Bittensor subnet 28. Capacity providers bid to serve inference, and that bidding sets the below-list price. Protocol data is public on Taostats.