Venice
Inference providervenice.ai ↗
Models on Venice · prices & measurements for models served by Venice’s APIUSD / MTOK · VERIFIED 4 MINUTES AGO
| MODEL | INPUT | OUTPUT | CONTEXT | QUANT | TTFT | TPS | SMOKE | VS BEST ELSEWHERE |
|---|---|---|---|---|---|---|---|---|
 GLM 4.7 FlashZ.aiBEST PRICE | $0.06 | $0.40 | 128K | fp8 | 0.99s° | 39° tok/s | — | cheapest here |
 Qwen3.6 35B A3BQwen | $0.10 | $1.00 | 256K | fp8 | 0.89s° | 242° tok/s | 1/5 | $0.07 elsewhere |
| Qwen3.5-9BQwenBEST PRICE | $0.10 | $0.15 | 256K | fp8 | 1.4s° | 149° tok/s | — | cheapest here |
 Gemma 4 31BGoogle | $0.12 | $0.36 | 256K | bf16 | 1.5s° | 81° tok/s | 5/5 | $0.08 elsewhere |
| Gemma 4 26B A4B Google | $0.13 | $0.40 | 256K | bf16 | 1.4s° | 47° tok/s | — | $0.042 elsewhere |
 DeepSeek V4 Flash 0423DeepSeek | $0.138 | $0.275 | 1M | — | 2.2s° | 120° tok/s | — | $0.056 elsewhere |
| Qwen3 235B A22B Instruct 2507Qwen | $0.15 | $0.75 | 128K | fp8 | 0.84s° | 30° tok/s | 5/5 | $0.087 elsewhere |
| DeepSeek V4 Flash 0731DeepSeek | $0.175 | $0.35 | 1M | — | 1.1s° | 76° tok/s | — | $0.04 elsewhere |
 Mistral Small 4Mistral | $0.188 | $0.75 | 256K | fp8 | 1s° | 61° tok/s | — | $0.15 elsewhere |
| CCUncensoredCognitive Computations | $0.20 | $0.90 | 128K | fp16 | 1.4s° | 100° tok/s | — | only provider |
| Qwen3 VL 235B A22B InstructQwen | $0.21 | $1.90 | 128K | fp8 | 1.1s° | 20° tok/s | — | $0.20 elsewhere |
 MiniMax M2.5MiniMaxBEST PRICE | $0.27 | $0.95 | 198K | — | 1.3s° | 59° tok/s | 2/5 | cheapest here |
| MiniMax M3MiniMax | $0.30 | $1.20 | 524K | fp8 | 0.82s° | 173° tok/s | — | $0.23 elsewhere |
| Qwen3.5-35B-A3BQwen | $0.313 | $1.25 | 256K | — | 1.2s° | 204° tok/s | — | $0.14 elsewhere |
| Qwen3.6 27BQwen | $0.325 | $3.25 | 256K | fp8 | 0.68s° | 30° tok/s | 2/5 | $0.30 elsewhere |
| DeepSeek V3.2DeepSeek | $0.33 | $0.48 | 160K | — | 2.3s° | 6° tok/s | — | $0.209 elsewhere |
| Qwen3 Coder 480B A35BQwen | $0.35 | $1.50 | 256K | fp8 | 0.65s° | 147° tok/s | — | $0.22 elsewhere |
| GLM 4.6Z.aiBEST PRICE | $0.43 | $1.75 | 198K | fp4 | 1.2s° | 12° tok/s | 3/5 | cheapest here |
| Qwen3.8 27BQwen | $0.45 | $3.20 | 262K | fp8 | 1.7s° | 93° tok/s | 4/5 | $0.35 elsewhere |
| Qwen3 235B A22B Thinking 2507Qwen | $0.45 | $3.50 | 128K | fp8 | 1.4s° | 40° tok/s | — | $0.23 elsewhere |
 Kimi K2.5Moonshot AI | $0.532 | $3.33 | 256K | — | 2.3s° | 446° tok/s | 3/5 | $0.45 elsewhere |
| GLM 4.7Z.ai | $0.55 | $2.65 | 198K | fp4 | 2.4s° | 27° tok/s | 3/5 | $0.40 elsewhere |
 Nemotron 3 UltraNVIDIA | $0.625 | $3.13 | 256K | fp8 | 3.6s° | — | — | $0.50 elsewhere |
| Kimi K2.7 CodeMoonshot AI | $0.75 | $3.50 | 256K | int4 | 0.89s° | 62° tok/s | 5/5 | $0.67 elsewhere |
| Kimi K2.6Moonshot AI | $0.75 | $3.50 | 256K | int4 | 1.4s° | 27° tok/s | — | $0.58 elsewhere |
| Qwen3.5 397B A17BQwen | $0.75 | $4.50 | 128K | — | 1.1s° | 108° tok/s | — | $0.39 elsewhere |
| GLM 5Z.ai | $1.00 | $3.20 | 198K | fp8 | 1.2s° | 122° tok/s | — | $0.60 elsewhere |
| GLM 5.2Z.ai | $1.40 | $4.40 | 1M | fp8 | 1.5s° | 77° tok/s | — | $0.50 elsewhere |
| GLM 5.1Z.ai | $1.54 | $4.84 | 200K | fp8 | 1s° | 21° tok/s | — | $0.91 elsewhere |
| DeepSeek V4 Pro 0423DeepSeek | $1.65 | $3.30 | 1M | — | 2s° | 55° tok/s | — | $0.522 elsewhere |
| Qwen3.8 2.4T A95BQwen | $2.50 | $7.50 | 262K | — | 1.5s° | 168° tok/s | 3/5 | $2.00 elsewhere |
TTFT/TPS: ModelIndex-measured medians (P50, trailing 72h) for Venice’s endpoints · ° = measured via OpenRouter routing · ✕ = probed, no measurable stream · SMOKE: of 5 capability probes passed · methodology on each model page