The Known Good Updated 25 Jul 2026
Providers· Inference host· 35 tracked models

SiliconFlow — pricing & performance

Compare models All providers Data sources
Models served
35
of 908 tracked models
Mean output speed
37
tokens per second, averaged over
34 measured endpoints
Mean latency
1.6s
time to first token, averaged over
34 measured endpoints
Cheapest blended
$0.075
per 1M tokens at a 3:1
input:output blend
Output speed on SiliconFlow

Median output tokens per second for each model this host serves, derived from OpenRouter endpoint telemetry. Higher is better. We do not run these measurements ourselves.

12 of 35 models ▾ ☰⚙⇩ +Add model from specific provider
116
102
77
72
68
62
50
50
49
48
44
41
The Known Good
35 models on SiliconFlow
Full leaderboard →
Model Creator Input $/M Output $/M Blended Speed TTFT Context
DeepSeek: DeepSeek V3 0324 DeepSeek $0.25 $1.00 $0.44 21 1.3s 164k
DeepSeek: DeepSeek V3.1 DeepSeek $0.27 $1.00 $0.45 19 1.6s 164k
DeepSeek: DeepSeek V3.1 Terminus DeepSeek $0.27 $1.00 $0.45 23 1.4s 164k
DeepSeek: DeepSeek V3.2 DeepSeek $0.26 $0.42 $0.30 19 1.8s 164k
DeepSeek: DeepSeek V3.2 Exp DeepSeek $0.27 $0.41 $0.30 14 1.7s 164k
DeepSeek: DeepSeek V4 Flash DeepSeek $0.13 $0.28 $0.17 72 1.1s 1.05M
DeepSeek: DeepSeek V4 Pro DeepSeek $1.50 $3.13 $1.91 41 2.1s 1.05M
DeepSeek: R1 0528 DeepSeek $0.50 $2.18 $0.92 23 1.3s 164k
Google: Gemma 4 26B A4B Google $0.12 $0.40 $0.19 23 1.6s 262k
Google: Gemma 4 31B Google $0.13 $0.40 $0.20 22 1.1s 262k
MiniMax: MiniMax M2.5 MiniMax $0.30 $1.20 $0.53 48 1.0s 197k
MoonshotAI: Kimi K2.5 Moonshot AI $0.45 $2.25 $0.90 50 1.5s 262k
MoonshotAI: Kimi K2.6 Moonshot AI $0.77 $3.40 $1.43 50 1.5s 262k
MoonshotAI: Kimi K2.7 Code Moonshot AI $0.86 $3.80 $1.59 38 1.3s 262k
Nex AGI: Nex-N2-Pro nex-agi $0.50 $2.50 $1.00 116 1.1s 262k
OpenAI: gpt-oss-120b OpenAI $0.050 $0.45 $0.15 26 1.3s 131k
OpenAI: gpt-oss-20b OpenAI $0.040 $0.18 $0.075 17 1.7s 131k
Qwen: Qwen3 30B A3B Instruct 2507 Alibaba $0.090 $0.30 $0.14 12 2.4s 262k
Qwen: Qwen3 32B Alibaba $0.14 $0.57 $0.25 28 1.5s 131k
Qwen: Qwen3 Coder 30B A3B Instruct Alibaba $0.070 $0.28 $0.12 23 1.4s 262k
Qwen: Qwen3 VL 30B A3B Instruct Alibaba $0.29 $1.00 $0.47 — — 262k
Qwen: Qwen3 VL 30B A3B Thinking Alibaba $0.29 $1.00 $0.47 77 1.1s 262k
Qwen: Qwen3.5-122B-A10B Alibaba $0.26 $2.08 $0.71 30 1.9s 262k
Qwen: Qwen3.5-27B Alibaba $0.25 $2.00 $0.69 11 1.3s 262k
Qwen: Qwen3.5-35B-A3B Alibaba $0.24 $1.80 $0.63 68 1.5s 262k
Qwen: Qwen3.5-9B Alibaba $0.10 $0.15 $0.11 18 2.3s 262k
Qwen: Qwen3.6 27B Alibaba $0.30 $3.20 $1.02 28 1.9s 262k
Qwen: Qwen3.6 35B A3B Alibaba $0.20 $1.60 $0.55 44 1.2s 262k
StepFun: Step 3.5 Flash stepfun $0.10 $0.30 $0.15 26 2.0s 262k
Tencent: Hunyuan A13B Instruct Tencent $0.14 $0.57 $0.25 3 1.1s 131k
Tencent: Hy3 preview Tencent $0.066 $0.26 $0.11 102 2.0s 262k
Z.ai: GLM 4.5 Air Z.ai $0.14 $0.86 $0.32 30 1.3s 131k
Z.ai: GLM 5 Z.ai $0.95 $2.55 $1.35 29 2.9s 205k
Z.ai: GLM 5.1 Z.ai $1.19 $3.74 $1.83 62 1.2s 205k
Z.ai: GLM 5.2 Z.ai $1.30 $4.09 $2.00 49 2.2s 1.05M
The Known Good
Source. Prices, endpoint availability and throughput for SiliconFlow are ingested from OpenRouter; model metadata is enriched from models.dev. A blank cell means the figure was not published for that endpoint — it never means zero. Blended price is (3 × input + output) ÷ 4, the same 3:1 blend used everywhere on this site; see methodology. Quality scores are not provider-specific: a model scores the same wherever it is hosted, so quality lives on the model pages.