HQ US · serves 14 tracked open-weights models. status page · privacy policy
- models
- 14
- verified
- —
- of overall top 10
- —
- cheapest on
- —
- vs best $
- 2.60×
- ctx (median)
- 256k
- uptime
- 97.9%
# models
| model | rank | $/M in | $/M out | ✓ |
|---|---|---|---|---|
| glm-4.7 | #28 | $0.60 | $2.20 | — |
| kimi-k2-thinking | #30 | $0.60 | $2.50 | — |
| deepseek-v3.2 | #32 | $0.56 | $1.68 | — |
| minimax-m2 | #46 | $0.30 | $1.20 | — |
| gemma-4-26b-a4b-it | #47 | $0.15 | $0.60 | — |
| qwen3-235b-a22b-instruct-2507 | #59 | $0.22 | $0.88 | — |
| qwen3-coder-480b-a35b | #60 | $0.22 | $1.80 | — |
| gpt-oss-120b | #63 | $0.09 | $0.36 | — |
| gpt-oss-20b | #75 | $0.07 | $0.25 | — |
| llama-4-maverick | #86 | $0.35 | $1.15 | — |
| llama-3.3-70b-instruct | #94 | $0.72 | $0.72 | — |
| llama-4-scout | #103 | $0.25 | $0.70 | — |
| deepseek-v4.1-flash | — | $0.30 | $1.20 | — |
| qwen3-next-80b-a3b-thinking | — | $0.15 | $1.20 | — |
all providers · machine-readable: providers.json