LLM API prices by model
20 models, each compared across every lab, inference host and router we track. Prices are USD per 1M tokens. Blended weights input and output 3:1, the usual chat workload.
Model labs 24Inference hosts 31Hyperscaler AI platforms 5LLM routers 14LLM gateways 12
| Model | Creator | Providers | Cheapest input | Cheapest output | Cheapest blended | Spread | Context |
|---|---|---|---|---|---|---|---|
| Llama 3.3 70B Instruct open meta-llama/llama-3.3-70b-instruct |
meta | 22 | $0.10 | $0.32 | $0.155 | 7.9× | 131k |
| Llama 4 Maverick 17B Instruct open meta-llama/llama-4-maverick |
meta | 12 | $0.188 | $0.652 | $0.304 | 2.1× | 1.05M |
| Llama 3.1 8B Instruct open meta-llama/llama-3.1-8b-instruct |
meta | 10 | $0.02 | $0.04 | $0.025 | 8.8× | 131k |
| Llama 4 Scout 17B Instruct open meta-llama/llama-4-scout |
meta | 10 | $0.08 | $0.29 | $0.14 | 3.0× | 10M |
| Llama 3.1 70B Instruct open meta-llama/llama-3.1-70b-instruct |
meta | 5 | $0.293 | $0.40 | $0.40 | 2.0× | 131k |
| Llama 3.2 1B Instruct open meta-llama/llama-3.2-1b-instruct |
meta | 4 | $0.027 | $0.10 | $0.07 | 1.4× | 128k |
| Llama 3.2 3B Instruct open meta-llama/llama-3.2-3b-instruct |
meta | 4 | $0.05 | $0.15 | $0.12 | 1.2× | 131k |
| Muse Glimmer 30B open meta/muse-glimmer-30b |
meta | 3 | $0.30 | $1.10 | $0.50 | 1.3× | 131k |
| Llama 3 8B Instruct Lite open meta-llama/llama-3-8b-instruct |
meta | 2 | $0.05 | $0.14 | $0.10 | 1.4× | — |
| Muse Spark 1.2 (Contributor) meta/muse-spark-1.2-contributor |
meta | 2 | $0.10 | $0.20 | $0.125 | 1.05M | |
| Muse Spark 1.3 (Contributor) meta/muse-spark-1.3-contributor |
meta | 2 | $0.10 | $0.20 | $0.125 | 1.05M | |
| Llama 3.2 90B Vision Instruct open meta-llama/llama-3.2-90b-vision-instruct |
meta | 2 | $0.35 | $0.40 | $0.362 | 2.0× | 131k |
| Muse Spark 1.1 meta/muse-spark-1.1 |
meta | 2 | $1.25 | $4.25 | $2 | 1.05M | |
| Muse Spark 1.2 meta/muse-spark-1.2 |
meta | 2 | $1.25 | $4.25 | $2 | 1.05M | |
| Muse Spark 1.3 meta/muse-spark-1.3 |
meta | 2 | $1.25 | $4.25 | $2 | 1.05M | |
| Llama 3.1 405B Instruct open meta-llama/llama-3.1-405b-instruct |
meta | 2 | — | — | — | 128k | |
| Meta: Muse Glimmer 30B (batch) meta/muse-glimmer-30b:batch |
meta | 1 | $0.175 | $0.75 | $0.319 | 131k | |
| Llama 3 70B Instruct open meta-llama/llama-3-70b-instruct |
meta | 1 | $0.65 | $2.75 | $1.18 | — | |
| Muse Image 1.0 meta/muse-image-1.0 |
meta | 1 | — | — | — | — | |
| Muse Voice Transcribe 1.0 meta/muse-voice-transcribe-1.0 |
meta | 1 | — | — | — | — |