DeepInfra vs Together AI: LLM API pricing compared
Updated 2026-11-10 · listed A to Z, not ranked
DeepInfra serves open models behind an OpenAI-compatible API. Language models are billed per token; some other models are billed by execution time. There are no long-term contracts.
Together AI hosts open-source and partner models behind a serverless API, rents GPU clusters, and runs fine-tuning jobs. Serverless inference bills per 1M tokens (with cheaper cached input on many models); GPUs bill per GPU hour.
Listed A to Z, from deepinfra.com and together.ai. Every figure below is quoted from those pages.
- Main price table
- DeepSeek-V4-Flash, Qwen3.8-Flash, gemma-4-31B-it-Ultra, DeepSeek-V4-Pro, GLM-5.3 (list price)
- Free option
- None in the published plan table
- Published prices
- DeepSeek-V4-Flash $0.09 in / $0.18 out, DeepSeek-V4-Pro $1.30 / $2.60, Qwen3.8-Flash $0.113 / $0.382.
- Main price table
- gpt-oss-120B, GLM-5.3, GLM-5.3-Flash, Kimi K3, DeepSeek V4.1 Flash, Llama 3 8B Instruct Lite
- Free option
- None in the published plan table
- Published prices
- Serverless tokens (e.g. gpt-oss-120B $0.15 / $0.60 per 1M), GPU clusters (H100 $3.99/hr on demand), dedicated inference and fine-tuning from $0.34 per 1M tokens.
How we checked: Prices quoted from each official pricing page: DeepInfra checked Nov 10, 2026, Together AI checked Nov 10, 2026. Taxes, usage overages and promotions can change what you pay; confirm with each provider before you buy.
How they compare
- Main price table. DeepInfra lists 5 entries (DeepSeek-V4-Flash, Qwen3.8-Flash, gemma-4-31B-it-Ultra, DeepSeek-V4-Pro, GLM-5.3 (list price)); Together AI lists 6 (gpt-oss-120B, GLM-5.3, GLM-5.3-Flash, Kimi K3, DeepSeek V4.1 Flash, Llama 3 8B Instruct Lite).
- Free option. DeepInfra: none in the published plan table. Together AI: none in the published plan table.
- How prices are quoted. DeepInfra: “Selected language models (per 1M tokens)”. Together AI: “Serverless inference: sample chat models (per 1M tokens)”.
DeepInfra plans and prices
| Model | Input | Cached input | Output |
|---|---|---|---|
| DeepSeek-V4-Flash | $0.09 | $0.018 | $0.18 |
| Qwen3.8-Flash | $0.113 | $0.0141 | $0.382 |
| gemma-4-31B-it-Ultra | $0.27 | – | $0.76 |
| DeepSeek-V4-Pro | $1.30 | $0.10 | $2.60 |
| GLM-5.3 (list price) | $0.90 | $0.20 | $4.00 |
GLM-5.3 was shown at 38% off its list price ($0.5625 input, $2.50 output) on Nov 10. Discounts change; check the page.
From deepinfra.com, checked Nov 10, 2026. Full breakdown: DeepInfra pricing.
Together AI plans and prices
| Model | Input | Cached input | Output |
|---|---|---|---|
| gpt-oss-120B | $0.15 | — | $0.60 |
| GLM-5.3 | $1.40 | $0.26 | $4.40 |
| GLM-5.3-Flash | $0.15 | $0.03 | $0.50 |
| Kimi K3 | $3.00 | $0.30 | $15.00 |
| DeepSeek V4.1 Flash | $0.30 | $0.006 | $1.20 |
| Llama 3 8B Instruct Lite | $0.14 | — | $0.14 |
Audio: Whisper Large v3 $0.0015 per audio minute; Cartesia Sonic-3 TTS $65.00 per 1M characters. Images from $0.0017 per megapixel (Juggernaut Lightning Flux).
| Hardware | Preemptible | On demand | Reserved 3–30 days | Reserved 91–180 days |
|---|---|---|---|---|
| NVIDIA HGX H100 | $1.99 | $3.99 | $3.69 | $3.19 |
| NVIDIA HGX H200 | $2.99 | $5.99 | $4.99 | $3.99 |
| NVIDIA HGX B200 | $4.09 | $8.19 | $7.99 | $6.79 |
| NVIDIA HGX B300 | $4.99 | $9.99 | Contact us | Contact us |
Dedicated model inference is listed separately: HGX H100 $5.49 and HGX B200 $8.99 per GPU hour on demand.
From together.ai, checked Nov 10, 2026. Full breakdown: Together AI pricing.
FAQ
How much does DeepInfra cost?
DeepSeek-V4-Flash $0.09 in / $0.18 out, DeepSeek-V4-Pro $1.30 / $2.60, Qwen3.8-Flash $0.113 / $0.382. (checked Nov 10, 2026).
How much does Together AI cost?
Serverless tokens (e.g. gpt-oss-120B $0.15 / $0.60 per 1M), GPU clusters (H100 $3.99/hr on demand), dedicated inference and fine-tuning from $0.34 per 1M tokens. (checked Nov 10, 2026).
Do DeepInfra and Together AI have a free option?
DeepInfra: none in the published plan table. Together AI: none in the published plan table.
Which is better, DeepInfra or Together AI?
We do not pick a winner. The two price and package their products differently; the tables above quote each official pricing page so you can match a plan to your own usage.
Where do these prices come from?
Each vendor's own pricing page: deepinfra.com (checked Nov 10, 2026) and together.ai (checked Nov 10, 2026). Prices change; confirm on the vendor's page before you buy.
Sources (official pages, read 2026-11-10)
Spot something out of date? Email hello@youraitools.co and we’ll fix it.