# Inferenceapis
serves 14 tracked open-weights models.
- models
- 14
- verified
- —
- of overall top 10
- 4
- cheapest on
- —
- vs best $
- 2.45×
- ctx (median)
- 1048k
# models
| model | rank | $/M in | $/M out | ✓ |
|---|---|---|---|---|
| glm-5.3 | #1 | $1.82 | $5.72 | — |
| glm-5.3-flash | #3 | $0.20 | $0.66 | — |
| deepseek-v4-flash-0731 | #6 | $0.19 | $0.38 | — |
| deepseek-v4-pro | #7 | $1.72 | $5.15 | — |
| minimax-m3 | #12 | $0.39 | $1.56 | — |
| kimi-k2.7-code | #19 | $0.89 | $4.42 | — |
| deepseek-v4-flash | #22 | $0.19 | $0.38 | — |
| gpt-oss-120b | #63 | $0.20 | $0.80 | — |
| qwen3-vl-235b-a22b-instruct | #73 | $0.26 | $1.15 | — |
| gpt-oss-20b | #75 | $0.04 | $0.19 | — |
| deepseek-v3 | #77 | $0.19 | $0.38 | — |
| llama-3.3-70b-instruct | #94 | $0.20 | $0.50 | — |
| deepseek-v4-pro-0813 | — | $1.72 | $5.15 | — |
| deepseek-v4.1-flash | — | $0.40 | $1.60 | — |
all providers · machine-readable: providers.json