Together AI API pricing
Checked 12 Sept 2026 LLMs
54 models from US. Blended cost runs from $0.075 to $6.00 per million tokens. LLM vendors run no affiliate programmes, so nothing on this page earns Pikkly anything.
Every model
Ranked by blended cost at 3:1 input to output.
| Source | ||||||
|---|---|---|---|---|---|---|
| gemma-3n-E4B-it Best value | $0.060 | $0.120 | — | 32,768 | $0.075 | ↗ |
| gpt-oss-20b | $0.050 | $0.200 | — | 131,072 | $0.087 | ↗ |
| qwen-2-1.5b-instruct | $0.100 | $0.100 | — | 32,768 | $0.100 | ↗ |
| together-ai-up-to-4b | $0.100 | $0.100 | — | — | $0.100 | ↗ |
| DeepSeek-V4-Flash-0731 | $0.140 | $0.280 | $0.030 | 1,048,576 | $0.175 | ↗ |
| Meta-Llama-3.1-8B-Instruct-Turbo | $0.180 | $0.180 | — | — | $0.180 | ↗ |
| Qwen3.5-9B | $0.170 | $0.250 | — | 262,144 | $0.190 | ↗ |
| together-ai-4.1b-8b | $0.200 | $0.200 | — | — | $0.200 | ↗ |
| Qwen3.8-Flash | $0.150 | $0.470 | — | 1,000,000 | $0.230 | ↗ |
| GLM-5.3-Flash | $0.150 | $0.500 | $0.030 | 1,048,575 | $0.237 | ↗ |
| gpt-oss-120b | $0.150 | $0.600 | — | 131,072 | $0.263 | ↗ |
| Llama-4-Scout-17B-16E-Instruct | $0.180 | $0.590 | — | — | $0.282 | ↗ |
| Qwen3-235B-A22B-fp8-tput | $0.200 | $0.600 | — | 40,000 | $0.300 | ↗ |
| together-ai-8.1b-21b | $0.300 | $0.300 | — | — | $0.300 | ↗ |
| Llama-4-Maverick-17B-128E-Instruct-FP8 | $0.270 | $0.850 | — | — | $0.415 | ↗ |
| GLM-4.5-Air-FP8 | $0.200 | $1.10 | — | 128,000 | $0.425 | ↗ |
| Qwen3-Next-80B-A3B-Instruct | $0.150 | $1.50 | — | 262,144 | $0.487 | ↗ |
| Qwen3-Next-80B-A3B-Thinking | $0.150 | $1.50 | — | 262,144 | $0.487 | ↗ |
| MiniMax-M3 | $0.300 | $1.20 | $0.060 | 524,288 | $0.525 | ↗ |
| gemma-4-31B-it | $0.390 | $0.970 | — | 262,144 | $0.535 | ↗ |
| Qwen3.7-Plus | $0.320 | $1.28 | — | 1,000,000 | $0.560 | ↗ |
| Mixtral-8x7B-Instruct-v0.1 | $0.600 | $0.600 | — | — | $0.600 | ↗ |
| Muse-Glimmer-30B | $0.350 | $1.50 | $0.040 | 131,072 | $0.637 | ↗ |
| Inkling-Small | $0.500 | $1.20 | $0.100 | 524,288 | $0.675 | ↗ |
| together-ai-21.1b-41b | $0.800 | $0.800 | — | — | $0.800 | ↗ |
| GLM-4.7 | $0.450 | $2.00 | — | 200,000 | $0.838 | ↗ |
| DeepSeek-V3.1 | $0.600 | $1.70 | — | 128,000 | $0.875 | ↗ |
| Meta-Llama-3.1-70B-Instruct-Turbo | $0.880 | $0.880 | — | — | $0.880 | ↗ |
| together-ai-41.1b-80b | $0.900 | $0.900 | — | — | $0.900 | ↗ |
| DeepSeek-R1-0528-tput | $0.550 | $2.19 | — | 128,000 | $0.960 | ↗ |
| GLM-4.6 | $0.600 | $2.20 | — | 200,000 | $1.00 | ↗ |
| Llama-3.3-70B-Instruct-Turbo | $1.04 | $1.04 | — | 131,072 | $1.04 | ↗ |
| Kimi-K2.5 | $0.500 | $2.80 | — | 256,000 | $1.08 | ↗ |
| Qwen3.6-Plus | $0.500 | $3.00 | — | 1,000,000 | $1.13 | ↗ |
| Qwen3-235B-A22B-Thinking-2507 | $0.650 | $3.00 | — | 256,000 | $1.24 | ↗ |
| DeepSeek-V3 | $1.25 | $1.25 | — | 65,536 | $1.25 | ↗ |
| nemotron-3-ultra-550b-a55b | $0.600 | $3.60 | $0.200 | 512,288 | $1.35 | ↗ |
| Qwen3.5-397B-A17B | $0.600 | $3.60 | $0.350 | 262,144 | $1.35 | ↗ |
| Kimi-K2-Instruct | $1.00 | $3.00 | — | — | $1.50 | ↗ |
| Kimi-K2-Instruct-0905 | $1.00 | $3.00 | — | 262,144 | $1.50 | ↗ |
| Qwen3-235B-A22B-Instruct-2507-tput | $0.200 | $6.00 | — | 262,000 | $1.65 | ↗ |
| Kimi-K2.7-Code | $0.950 | $4.00 | $0.190 | 262,144 | $1.71 | ↗ |
| Inkling | $1.00 | $4.05 | $0.170 | 524,288 | $1.76 | ↗ |
| together-ai-81.1b-110b | $1.80 | $1.80 | — | — | $1.80 | ↗ |
| DeepSeek-V4-Pro-0813 | $1.32 | $3.96 | $0.130 | 1,048,576 | $1.98 | ↗ |
| Qwen3-Coder-480B-A35B-Instruct-FP8 | $2.00 | $2.00 | — | 256,000 | $2.00 | ↗ |
| GLM-5.2 | $1.40 | $4.40 | $0.260 | 1,048,575 | $2.15 | ↗ |
| GLM-5.3 | $1.40 | $4.40 | $0.260 | 1,048,575 | $2.15 | ↗ |
| DeepSeek-V4-Pro | $1.74 | $3.48 | $0.200 | 512,000 | $2.18 | ↗ |
| Qwen3.8-2.4T-A95B | $2.00 | $6.00 | $0.250 | 1,010,000 | $3.00 | ↗ |
| Meta-Llama-3.1-405B-Instruct-Turbo | $3.50 | $3.50 | — | — | $3.50 | ↗ |
| Qwen3.7-Max | $2.50 | $7.50 | $0.500 | 1,000,000 | $3.75 | ↗ |
| DeepSeek-R1 | $3.00 | $7.00 | — | 128,000 | $4.00 | ↗ |
| Kimi-K3 | $3.00 | $15.00 | $0.300 | 1,048,576 | $6.00 | ↗ |
Last checked . Methodology
All models · Llama-4-Scout-17B-16E-Instruct · Llama-4-Maverick-17B-128E-Instruct-FP8 · DeepSeek-V3.1 · Llama-3.3-70B-Instruct-Turbo · DeepSeek-V3 · Run your own on a GPU