CHANNEL PRICES LLM API Channel Prices The same model at different providers: reseller and independent-compute prices side by side, in USD per million tokens, with official list price for reference.
Data updated: 2026-11-05 01:31 UTC Jump to DeepSeek V4.1 Flash17 DeepSeek V4 Pro17 Qwen3.8 27B15 Qwen3.8 2.4T A95B4 GLM-5.3-Flash15 GLM-5.315 GLM-5.215 GLM-58 GLM-5.112 Kimi K316 Kimi K2.55 Kimi K2.615 Kimi K2.7 Code9 Hunyuan A13B1 MiniMax M313 MiniMax M3.1 Flash12 Step 5 Preview2 Claude Sonnet 51 CHANNEL DETAIL
DeepSeek V4.1 FlashDeepSeek · official blended price $0.53 Provider / plan Channel type Compute origin Blended USD/1M Verified Source StreamLake China fp8 · 1024K context Independent GPU provider Self-hosted $0.10 2026-11-05 Source DeepInfra US fp8 · 1049K context Independent GPU provider Self-hosted $0.11 2026-11-05 Source GMICloud US fp8 · 1049K context Independent GPU provider Self-hosted $0.11 2026-11-05 Source Venice US 1000K context Independent GPU provider Self-hosted $0.12 2026-11-05 Source DigitalOcean US 1049K context Overseas cloud reseller Self-hosted $0.12 2026-11-05 Source Fireworks US 1049K context Independent GPU provider Self-hosted $0.13 2026-11-05 Source Relace fp4 · 1049K context Independent GPU provider Self-hosted $0.15 2026-11-05 Source SiliconFlow Singapore fp8 · 1049K context Inference platform Self-hosted $0.17 2026-11-05 Source Baidu China fp8 · 1049K context China cloud reseller Self-hosted $0.17 2026-11-05 Source AtlasCloud US fp4 · 1049K context Independent GPU provider Self-hosted $0.17 2026-11-05 Source Parasail US fp8 · 1049K context Independent GPU provider Self-hosted $0.17 2026-11-05 Source Alibaba Singapore fp8 · 1000K context China cloud reseller Self-hosted $0.19 2026-11-05 Source Alibaba Cloud Bailian China DeepSeek-V4.1-Flash North China 2 (Beijing) China cloud reseller Self-hosted $0.26–$0.52 Off-peak hours 22:00–08:00 the next day $0.26 Peak hours 08:00–22:00 $0.52 2026-10-19 Source Mancer 2 fp8 · 1049K context Independent GPU provider Self-hosted $0.27 2026-11-05 Source Azure US 1049K context Overseas cloud reseller Self-hosted $0.30 2026-11-05 Source SiliconFlow China DeepSeek-V4-Flash Inference platform Self-hosted $0.34–$0.67 02:00–08:00 $0.34 00:00–02:00 and 08:00–24:00 $0.67 2026-09-03 Source Wafer US mxfp4 · 1049K context Independent GPU provider Self-hosted $0.49 2026-11-05 Source
CHANNEL DETAIL
DeepSeek V4 ProDeepSeek · official blended price $1.98 Provider / plan Channel type Compute origin Blended USD/1M Verified Source Relace fp4 · 1049K context Independent GPU provider Self-hosted $1.05 2026-11-05 Source StreamLake China fp8 · 1024K context Independent GPU provider Self-hosted $1.19 2026-11-05 Source GMICloud US fp8 · 1049K context Independent GPU provider Self-hosted $1.20 2026-11-05 Source DigitalOcean US 1049K context Overseas cloud reseller Self-hosted $1.30 2026-11-05 Source SiliconFlow Singapore fp8 · 1049K context Inference platform Self-hosted $1.53 2026-11-05 Source DeepInfra US fp8 · 1049K context Independent GPU provider Self-hosted $1.62 2026-11-05 Source Alibaba Singapore fp8 · 1000K context China cloud reseller Self-hosted $1.77 2026-11-05 Source Alibaba Singapore fp8 · 1000K context China cloud reseller Self-hosted $1.95 2026-11-05 Source Novita US fp8 · 1049K context Independent GPU provider Self-hosted $2.00 2026-11-05 Source Venice US 1000K context Independent GPU provider Self-hosted $2.06 2026-11-05 Source AtlasCloud US fp4 · 1049K context Independent GPU provider Self-hosted $2.11 2026-11-05 Source Baidu China fp8 · 1049K context China cloud reseller Self-hosted $2.11 2026-11-05 Source AWS Bedrock US DeepSeek-V4-Pro regular price Overseas cloud reseller Self-hosted $2.17 2026-11-05 Source Nebius Netherlands fp8 · 1049K context Independent GPU provider Self-hosted $2.19 2026-11-05 Source Alibaba Cloud Bailian China DeepSeek-V4-Pro North China 2 (Beijing) China cloud reseller Self-hosted $2.24 2026-11-05 Source SiliconFlow China DeepSeek-V4-Pro Inference platform Self-hosted $2.24 2026-11-05 Source Azure US 1049K context Overseas cloud reseller Self-hosted $2.39 2026-11-05 Source
CHANNEL DETAIL
Qwen3.8 27BAlibaba (Qwen) · official blended price $1.13 Provider / plan Channel type Compute origin Blended USD/1M Verified Source PrimeIntellect fp8 · 131K context Independent GPU provider Self-hosted $0.20 2026-11-05 Source Chutes US fp8 · 262K context Independent GPU provider Self-hosted $0.34 2026-11-05 Source Near AI fp8 · 262K context Independent GPU provider Self-hosted $0.37 2026-11-05 Source Darkbloom fp4 · 262K context Independent GPU provider Self-hosted $0.40 2026-11-05 Source StreamLake China 1024K context Independent GPU provider Self-hosted $0.45 2026-11-05 Source Ionstream US fp8 · 262K context Independent GPU provider Self-hosted $0.52 2026-11-05 Source DeepInfra US bf16 · 262K context Independent GPU provider Self-hosted $0.58 2026-11-05 Source Modal US 262K context Independent GPU provider Self-hosted $0.60 2026-11-05 Source Tenstorrent US fp4 · 262K context Independent GPU provider Self-hosted $0.61 2026-11-05 Source Wafer US fp8 · 262K context Independent GPU provider Self-hosted $0.62 2026-11-05 Source DekaLLM Indonesia 262K context Independent GPU provider Self-hosted $0.66 2026-11-05 Source AkashML fp8 · 262K context Independent GPU provider Self-hosted $0.66 2026-11-05 Source Parasail US fp8 · 262K context Independent GPU provider Self-hosted $0.73 2026-11-05 Source Reka fp8 · 262K context Independent GPU provider Self-hosted $0.74 2026-11-05 Source Mancer 2 fp8 · 262K context Independent GPU provider Self-hosted $0.76 2026-11-05 Source
CHANNEL DETAIL
Qwen3.8 2.4T A95BAlibaba (Qwen) · official blended price $3.00 Provider / plan Channel type Compute origin Blended USD/1M Verified Source Novita US 1000K context Independent GPU provider Self-hosted $3.00 2026-11-05 Source SiliconFlow Singapore fp8 · 1049K context Inference platform Self-hosted $3.00 2026-11-05 Source Venice US 262K context Independent GPU provider Self-hosted $3.00 2026-11-05 Source Together US 1010K context Independent GPU provider Self-hosted $3.00 2026-11-05 Source
CHANNEL DETAIL
GLM-5.3-FlashZhipu AI · official blended price $0.24 Provider / plan Channel type Compute origin Blended USD/1M Verified Source Relace fp4 · 1049K context Independent GPU provider Self-hosted $0.04 2026-11-05 Source OpenInference US fp4 · 1049K context Independent GPU provider Self-hosted $0.08 2026-11-05 Source Wafer US nvfp4 · 1049K context Independent GPU provider Self-hosted $0.08 2026-11-05 Source Wafer US nvfp4 · 1049K context Independent GPU provider Self-hosted $0.09 2026-11-05 Source Darkbloom fp4 · 262K context Independent GPU provider Self-hosted $0.09 2026-11-05 Source Morph US fp8 · 1049K context Independent GPU provider Self-hosted $0.11 2026-11-05 Source Tenstorrent US fp8 · 262K context Independent GPU provider Self-hosted $0.12 2026-11-05 Source GMICloud US fp8 · 1049K context Independent GPU provider Self-hosted $0.12 2026-11-05 Source Novita US fp8 · 1049K context Independent GPU provider Self-hosted $0.13 2026-11-05 Source StreamLake China fp8 · 1024K context Independent GPU provider Self-hosted $0.14 2026-11-05 Source Reka 1049K context Independent GPU provider Self-hosted $0.17 2026-11-05 Source Makora fp4 · 1049K context Independent GPU provider Self-hosted $0.17 2026-11-05 Source InferenceNet US fp4 · 1049K context Independent GPU provider Self-hosted $0.17 2026-11-05 Source InferenceNet US fp4 · 1049K context Independent GPU provider Self-hosted $0.17 2026-11-05 Source Phala US nvfp4 · 1049K context Independent GPU provider Self-hosted $0.18 2026-11-05 Source
CHANNEL DETAIL
GLM-5.3Zhipu AI · official blended price $2.15 Provider / plan Channel type Compute origin Blended USD/1M Verified Source Wafer US fp4 · 1049K context Independent GPU provider Self-hosted $0.40 2026-11-05 Source Baidu China fp8 · 1049K context China cloud reseller Self-hosted $0.43 2026-11-05 Source Wafer US fp4 · 1049K context Independent GPU provider Self-hosted $0.48 2026-11-05 Source Reka 1049K context Independent GPU provider Self-hosted $0.63 2026-11-05 Source Morph US fp8 · 1049K context Independent GPU provider Self-hosted $0.68 2026-11-05 Source Decart US fp4 · 1049K context Independent GPU provider Self-hosted $0.86 2026-11-05 Source DekaLLM Indonesia 1049K context Independent GPU provider Self-hosted $0.87 2026-11-05 Source Makora fp4 · 1049K context Independent GPU provider Self-hosted $0.88 2026-11-05 Source InferenceNet US fp4 · 1049K context Independent GPU provider Self-hosted $0.96 2026-11-05 Source InferenceNet US fp4 · 1049K context Independent GPU provider Self-hosted $0.96 2026-11-05 Source Novita US fp8 · 1049K context Independent GPU provider Self-hosted $0.97 2026-11-05 Source DeepInfra US fp4 · 1049K context Independent GPU provider Self-hosted $1.05 2026-11-05 Source SiliconFlow Singapore fp8 · 1049K context Inference platform Self-hosted $1.07 2026-11-05 Source GMICloud US fp8 · 1049K context Independent GPU provider Self-hosted $1.07 2026-11-05 Source Cloudflare US 1049K context Independent GPU provider Self-hosted $1.07 2026-11-05 Source
CHANNEL DETAIL
GLM-5.2Zhipu AI · official blended price $2.15 Provider / plan Channel type Compute origin Blended USD/1M Verified Source Morph US fp8 · 1049K context Independent GPU provider Self-hosted $0.68 2026-11-05 Source DeepInfra US fp4 · 1049K context Independent GPU provider Self-hosted $0.70 2026-11-05 Source DeepInfra US fp4 · 1049K context Independent GPU provider Self-hosted $0.87 2026-11-05 Source StreamLake China fp8 · 1024K context Independent GPU provider Self-hosted $0.94 2026-11-05 Source Ambient fp8 · 203K context Independent GPU provider Self-hosted $0.95 2026-11-05 Source Novita US fp8 · 1049K context Independent GPU provider Self-hosted $1.00 2026-11-05 Source CoreWeave US fp4 · 1049K context Independent GPU provider Self-hosted $1.07 2026-11-05 Source DigitalOcean US 1049K context Overseas cloud reseller Self-hosted $1.07 2026-11-05 Source SiliconFlow Singapore fp8 · 1049K context Inference platform Self-hosted $1.07 2026-11-05 Source AtlasCloud US fp8 · 1049K context Independent GPU provider Self-hosted $1.44 2026-11-05 Source Alibaba Singapore fp8 · 1000K context China cloud reseller Self-hosted $1.48 2026-11-05 Source Wafer US fp4 · 1049K context Independent GPU provider Self-hosted $1.70 2026-11-05 Source Alibaba Singapore fp8 · 1000K context China cloud reseller Self-hosted $1.72 2026-11-05 Source Relace fp4 · 1049K context Independent GPU provider Self-hosted $1.98 2026-11-05 Source Wafer US fp4 · 1049K context Independent GPU provider Self-hosted $2.14 2026-11-05 Source
CHANNEL DETAIL
GLM-5Zhipu AI · official blended price $1.55 Provider / plan Channel type Compute origin Blended USD/1M Verified Source StreamLake China fp8 · 198K context Independent GPU provider Self-hosted $0.93 2026-11-05 Source GMICloud US fp8 · 203K context Independent GPU provider Self-hosted $0.93 2026-11-05 Source Baidu China fp8 · 203K context China cloud reseller Self-hosted $1.09 2026-11-05 Source SiliconFlow Singapore fp8 · 205K context Inference platform Self-hosted $1.35 2026-11-05 Source Amazon Bedrock US 203K context Overseas cloud reseller Self-hosted $1.55 2026-11-05 Source Venice US fp8 · 198K context Independent GPU provider Self-hosted $1.55 2026-11-05 Source Novita US fp8 · 203K context Independent GPU provider Self-hosted $1.55 2026-11-05 Source Venice US 200K context Independent GPU provider Self-hosted $1.90 2026-11-05 Source
CHANNEL DETAIL
GLM-5.1Zhipu AI · official blended price $2.15 Provider / plan Channel type Compute origin Blended USD/1M Verified Source Chutes US fp8 · 203K context Independent GPU provider Self-hosted $1.40 2026-11-05 Source StreamLake China fp8 · 200K context Independent GPU provider Self-hosted $1.48 2026-11-05 Source SiliconFlow Singapore fp8 · 205K context Inference platform Self-hosted $1.83 2026-11-05 Source AtlasCloud US fp8 · 203K context Independent GPU provider Self-hosted $1.94 2026-11-05 Source Phala US 203K context Independent GPU provider Self-hosted $1.96 2026-11-05 Source Alibaba Singapore fp8 · 203K context China cloud reseller Self-hosted $2.04 2026-11-05 Source Novita US fp8 · 205K context Independent GPU provider Self-hosted $2.14 2026-11-05 Source Nebius Netherlands fp8 · 203K context Independent GPU provider Self-hosted $2.15 2026-11-05 Source Baidu China fp8 · 203K context China cloud reseller Self-hosted $2.15 2026-11-05 Source GMICloud US fp8 · 203K context Independent GPU provider Self-hosted $2.15 2026-11-05 Source Friendli US 203K context Independent GPU provider Self-hosted $2.15 2026-11-05 Source Venice US fp8 · 200K context Independent GPU provider Self-hosted $2.15 2026-11-05 Source
CHANNEL DETAIL
Kimi K3Moonshot AI · official blended price $6.00 Provider / plan Channel type Compute origin Blended USD/1M Verified Source Wafer US mxfp4 · 1049K context Independent GPU provider Self-hosted $2.61 2026-11-05 Source Relace fp4 · 1049K context Independent GPU provider Self-hosted $2.76 2026-11-05 Source Makora fp4 · 1049K context Independent GPU provider Self-hosted $3.27 2026-11-05 Source ParetoAlpha US 1049K context Independent GPU provider Self-hosted $3.56 2026-11-05 Source Chutes US mxfp4 · 1049K context Independent GPU provider Self-hosted $3.76 2026-11-05 Source Decart US mxfp4 · 1049K context Independent GPU provider Self-hosted $3.90 2026-11-05 Source Wafer US mxfp4 · 1049K context Independent GPU provider Self-hosted $4.14 2026-11-05 Source Fireworks US 1049K context Independent GPU provider Self-hosted $4.20 2026-11-05 Source InferenceNet US fp4 · 1049K context Independent GPU provider Self-hosted $4.30 2026-11-05 Source InferenceNet US fp4 · 1049K context Independent GPU provider Self-hosted $4.30 2026-11-05 Source Morph US fp8 · 1049K context Independent GPU provider Self-hosted $4.59 2026-11-05 Source Phala US 1049K context Independent GPU provider Self-hosted $5.10 2026-11-05 Source DigitalOcean US 1049K context Overseas cloud reseller Self-hosted $5.15 2026-11-05 Source Io Net US fp4 · 1048K context Independent GPU provider Self-hosted $5.38 2026-11-05 Source SiliconFlow China Kimi-K3 Inference platform Self-hosted $5.40 2026-11-05 Source Wafer US mxfp4 · 1049K context Independent GPU provider Self-hosted $5.61 2026-11-05 Source
CHANNEL DETAIL
Kimi K2.5Moonshot AI · official blended price $1.20 Provider / plan Channel type Compute origin Blended USD/1M Verified Source AtlasCloud US int4 · 262K context Independent GPU provider Self-hosted $0.99 2026-11-05 Source SiliconFlow Singapore int4 · 262K context Inference platform Self-hosted $1.00 2026-11-05 Source Novita US 262K context Independent GPU provider Self-hosted $1.14 2026-11-05 Source Amazon Bedrock US 262K context Overseas cloud reseller Self-hosted $1.20 2026-11-05 Source Venice US 256K context Independent GPU provider Self-hosted $1.23 2026-11-05 Source
CHANNEL DETAIL
Kimi K2.6Moonshot AI · official blended price $1.71 Provider / plan Channel type Compute origin Blended USD/1M Verified Source Chutes US int4 · 262K context Independent GPU provider Self-hosted $0.74 2026-11-05 Source DigitalOcean US 262K context Overseas cloud reseller Self-hosted $1.03 2026-11-05 Source StreamLake China fp8 · 256K context Independent GPU provider Self-hosted $1.08 2026-11-05 Source CoreWeave US fp4 · 262K context Independent GPU provider Self-hosted $1.34 2026-11-05 Source Crusoe US bf16 · 262K context Independent GPU provider Self-hosted $1.40 2026-11-05 Source SiliconFlow Singapore fp8 · 262K context Inference platform Self-hosted $1.43 2026-11-05 Source DeepInfra US fp4 · 262K context Independent GPU provider Self-hosted $1.44 2026-11-05 Source Venice US int4 · 256K context Independent GPU provider Self-hosted $1.44 2026-11-05 Source Parasail US int4 · 262K context Independent GPU provider Self-hosted $1.44 2026-11-05 Source Novita US 262K context Independent GPU provider Self-hosted $1.45 2026-11-05 Source Inceptron Sweden int4 · 262K context Independent GPU provider Self-hosted $1.54 2026-11-05 Source GMICloud US fp8 · 262K context Independent GPU provider Self-hosted $1.54 2026-11-05 Source Baidu China fp4 · 262K context China cloud reseller Self-hosted $1.71 2026-11-05 Source AtlasCloud US int4 · 262K context Independent GPU provider Self-hosted $1.71 2026-11-05 Source Cloudflare US 262K context Independent GPU provider Self-hosted $1.71 2026-11-05 Source
CHANNEL DETAIL
Kimi K2.7 CodeMoonshot AI · official blended price $1.71 Provider / plan Channel type Compute origin Blended USD/1M Verified Source CoreWeave US int4 · 262K context Independent GPU provider Self-hosted $1.20 2026-11-05 Source StreamLake China 256K context Independent GPU provider Self-hosted $1.28 2026-11-05 Source Inceptron Sweden int4 · 262K context Independent GPU provider Self-hosted $1.32 2026-11-05 Source Venice US int4 · 256K context Independent GPU provider Self-hosted $1.44 2026-11-05 Source SiliconFlow Singapore fp8 · 262K context Inference platform Self-hosted $1.59 2026-11-05 Source Novita US int4 · 262K context Independent GPU provider Self-hosted $1.64 2026-11-05 Source GMICloud US fp8 · 262K context Independent GPU provider Self-hosted $1.71 2026-11-05 Source Alibaba Singapore fp8 · 262K context China cloud reseller Self-hosted $1.71 2026-11-05 Source Cloudflare US 262K context Independent GPU provider Self-hosted $1.71 2026-11-05 Source
CHANNEL DETAIL
Hunyuan A13BTencent · official blended price $0.13 Provider / plan Channel type Compute origin Blended USD/1M Verified Source SiliconFlow Singapore fp8 · 131K context Inference platform Self-hosted $0.25 2026-11-05 Source
CHANNEL DETAIL
MiniMax M3MiniMax · official blended price $0.53 Provider / plan Channel type Compute origin Blended USD/1M Verified Source DeepInfra US fp8 · 524K context Independent GPU provider Self-hosted $0.39 2026-11-05 Source CoreWeave US fp4 · 262K context Independent GPU provider Self-hosted $0.41 2026-11-05 Source GMICloud US fp8 · 1049K context Independent GPU provider Self-hosted $0.42 2026-11-05 Source DeepInfra US fp8 · 524K context Independent GPU provider Self-hosted $0.49 2026-11-05 Source SiliconFlow China MiniMax-M3 Inference platform Self-hosted $0.52 2026-11-05 Source StreamLake China fp8 · 1000K context Independent GPU provider Self-hosted $0.52 2026-11-05 Source Venice US fp8 · 524K context Independent GPU provider Self-hosted $0.52 2026-11-05 Source Together US 524K context Independent GPU provider Self-hosted $0.52 2026-11-05 Source Parasail US fp8 · 1049K context Independent GPU provider Self-hosted $0.52 2026-11-05 Source AtlasCloud US fp8 · 524K context Independent GPU provider Self-hosted $0.52 2026-11-05 Source Novita US fp8 · 1000K context Independent GPU provider Self-hosted $0.52 2026-11-05 Source ModelRun US fp8 · 1049K context Independent GPU provider Self-hosted $0.55 2026-11-05 Source Mara US 1049K context Independent GPU provider Self-hosted $1.05 2026-11-05 Source
CHANNEL DETAIL
MiniMax M3.1 FlashMiniMax · official blended price $0.26 Provider / plan Channel type Compute origin Blended USD/1M Verified Source GMICloud US fp8 · 1049K context Independent GPU provider Self-hosted $0.20 2026-11-05 Source StreamLake China fp8 · 1000K context Independent GPU provider Self-hosted $0.24 2026-11-05 Source Fireworks US 1049K context Independent GPU provider Self-hosted $0.26 2026-11-05 Source InferenceNet US fp4 · 1000K context Independent GPU provider Self-hosted $0.26 2026-11-05 Source DigitalOcean US 1049K context Overseas cloud reseller Self-hosted $0.26 2026-11-05 Source Novita US fp16, 1000K context Independent GPU provider Self-hosted $0.26 2026-11-05 Source Nebius Netherlands fp4 · 1049K context Independent GPU provider Self-hosted $0.26 2026-11-05 Source SambaNova US 1049K context Independent GPU provider Self-hosted $0.26 2026-11-05 Source ModelRun US fp4 · 1049K context Independent GPU provider Self-hosted $0.26 2026-11-05 Source Parasail US fp8 · 1049K context Independent GPU provider Self-hosted $0.26 2026-11-05 Source InferenceNet US fp4 · 1000K context Independent GPU provider Self-hosted $0.26 2026-11-05 Source Together US 1049K context Independent GPU provider Self-hosted $0.26 2026-11-05 Source
CHANNEL DETAIL
Step 5 PreviewStepFun · official blended price $1.53 Provider / plan Channel type Compute origin Blended USD/1M Verified Source GMICloud US fp8 · 1049K context Independent GPU provider Self-hosted $1.43 2026-11-05 Source Featherless US fp4 · 262K context Independent GPU provider Self-hosted $1.43 2026-11-05 Source
CHANNEL DETAIL
Claude Sonnet 5Anthropic · official blended price $4.00 Provider / plan Channel type Compute origin Blended USD/1M Verified Source AWS Bedrock US Claude Sonnet 5 on-demand Overseas cloud reseller Official API resale $4.00 2026-11-05 Source
What do the two kinds of “Compute origin” mean? Official API resale : the provider does not host the model and only resells access to the vendor’s own API, so prices track the official list price. Self-hosted open-weights model : the weights are open and the provider runs its own GPUs, so the price reflects its own compute cost.
Why do open-weights models have so many channels? Open-weights models (DeepSeek, Kimi, MiniMax, GLM, Hunyuan and others) naturally have many independent compute providers competing; closed models (GPT, Claude, Gemini) can only be resold.
How is the blended price calculated, and where does the data come from? Blended price = (3 × input + 1 × output) / 4, converted from CNY at ¥6.68 per USD. Open-model channel data comes from the public OpenRouter API (fetched daily); other channels are re-checked daily against their pricing pages.
Can I buy at these prices? For comparison only; not a price quote, and TANDIU does not resell.