Providers /Novita AI
Inference provider · HQ SG

Novita AI

Models
149
Regions
00

Our take

Editorial by LLMap · updated Aug 3, 2026

Novita AI is a Singapore-based host with the widest catalogue of downloadable models we track and strong privacy guarantees including a zero-retention option. Its 149 offers give maximum choice, though throughput data is sparse and compliance documentation is thinner than some rivals.

Use this when you need the broadest model choice available through a single provider. Pick it for privacy-sensitive workloads that benefit from a zero-retention option and a no-training-on-prompts commitment. Consider it for cost-probing on small or edge models, where entry prices sit at the bottom of the range we sample. Skip it if you require verified SOC 2, a disclosed EU endpoint, or predictable throughput across most of the catalogue.

Strengths
  • Deepest catalogue of any provider we track — 149 offers.
  • Strongest privacy guarantees among sampled providers: no training on prompts, with a zero-retention option available.
  • Lowest entry price in sampled offers on inclusionai-ling-2-6-flash.
  • Highest disclosed throughput in the sample set: openai-gpt-oss-safeguard-20b at 84 tokens per second.
Trade-offs
  • Compliance and regional presence are thinly documented: SOC 2 is unverified in our data and no EU endpoint is disclosed.
  • Throughput data is mostly missing, making performance unpredictable — 8 of 12 sample offers show no measured tokens-per-second figure.
  • Listing quality concerns: meta-llama-llama-3-1-8b-instruct appears twice at identical prices with different throughput states, one measured and one not.
01

Privacy & data handling

Yes
No
No answer on record

The colour says whether the answer favours you, not whether it is a yes — not training on your prompts earns a green cross, no zero-retention option earns a red one.

Trains on your prompts
No
Logs prompts
No
Offered
No answer on record

Prompt retention: none.

02

Models & pricing

DeepSeek V3685B$0.40 / $1.30 /1M64K11m agoDeepSeek V3685B$0.89 / $0.89 /1M64K6h agoDeepSeek V3 0324685B$0.27 / $1.12 /1M164K6h agoDeepSeek V3 0324685B$0.27 / $1.12 /1M164K11m agoDeepSeek V3.1 Terminus685B$0.27 / $1.00 /1M131K6h agoDeepSeek V3.1 Terminus685B$0.27 / $1.00 /1M131K11m agoDeepSeek V3.2685B$0.27 / $0.40 /1M164K11m agoDeepSeek V3.2685B$0.27 / $0.40 /1M164K6h agoDeepSeek V3.2 Exp685B$0.27 / $0.41 /1M164K11m agoDeepSeek V3.2 Exp685B$0.27 / $0.41 /1M164K6h agoDeepSeek V4 Flash291B$0.14 / $0.28 /1M1M11m agoDeepSeek V4 Flash291B$0.14 / $0.28 /1M1M6h agoDeepSeek V4 Pro1.6T$1.17 / $2.34 /1M1M11m agoDeepSeek V4 Pro1.6T$1.60 / $3.20 /1M1M6h agoERNIE 4.5 VL 424B A47B424B$0.42 / $1.25 /1M123K6h agoERNIE 4.5 VL 424B A47B424B$0.42 / $1.25 /1M123K11m agoGemma 3 12B12.2B$0.050 / $0.10 /1M131K6h agoGemma 3 27B27.4B$0.12 / $0.20 /1M98K6h agoGemma 3 27B27.4B$0.12 / $0.20 /1M98K11m agoGemma 4 26B A4B26.5B$0.13 / $0.40 /1M262K6h agoGemma 4 26B A4B26.5B$0.13 / $0.40 /1M262K11m agoGemma 4 31B32.7B$0.14 / $0.40 /1M262K11m agoGemma 4 31B32.7B$0.14 / $0.40 /1M262K6h agoGLM 4.5358B$0.60 / $2.20 /1M131K6h agoGLM 4.5 Air111B$0.13 / $0.85 /1M131K9m agoGLM 4.5 Air111B$0.13 / $0.85 /1M131K6h agoGLM 4.5V108B$0.60 / $1.80 /1M66K6h agoGLM 4.5V108B$0.60 / $1.80 /1M66K9m agoGLM 4.6357B$0.55 / $2.20 /1M205K6h agoGLM 4.6357B$0.55 / $2.20 /1M205K6h agoGLM 4.6V108B$0.30 / $0.90 /1M131K6h agoGLM 4.6V108B$0.30 / $0.90 /1M131K14h agoGLM 4.7358B$0.60 / $2.20 /1M205K6h agoGLM 4.7358B$0.54 / $1.98 /1M205K6h agoGLM 4.7 Flash31.2B$0.070 / $0.40 /1M200K6h agoGLM 4.7 Flash31.2B$0.070 / $0.40 /1M200K6h agoGLM 5754B$1.00 / $3.20 /1M203K9m agoGLM 5754B$1.00 / $3.20 /1M203K6h agoGLM 5.1754B$1.38 / $4.40 /1M205K6h agoGLM 5.1754B$1.38 / $4.40 /1M205K9m agoGLM 5.2753B$1.40 / $4.40 /1M1M6h agoGLM 5.2753B$0.63 / $1.98 /1M1M9m agoGLM 5 Turbo$1.20 / $4.00 /1M203K6h agoGLM 5V Turbo$1.20 / $4.00 /1M205K6h agogpt-oss-120b120B$0.050 / $0.25 /1M131K10m agogpt-oss-120b120B$0.050 / $0.25 /1M131K6h agogpt-oss-safeguard-20b21.5B$0.040 / $0.15 /1M131K6h agogpt-oss-safeguard-20b21.5B$0.040 / $0.15 /1M131K10m agoHy3299B$0.14 / $0.58 /1M262K9m agoHy3299B$0.14 / $0.58 /1M262K6h agoKimi K2 07111T$0.57 / $2.30 /1M131K6h agoKimi K2 07111T$0.57 / $2.30 /1M131K11m agoKimi K2 09051T$0.60 / $2.50 /1M262K6h agoKimi K2 09051T$0.60 / $2.50 /1M262K11m agoKimi K2.51.1T$0.60 / $3.00 /1M262K6h agoKimi K2.51.1T$0.57 / $2.85 /1M262K11m agoKimi K2.61.1T$0.80 / $3.40 /1M262K11m agoKimi K2.61.1T$0.80 / $3.40 /1M262K6h agoKimi K2.7 Code1.1T$0.91 / $3.84 /1M262K11m agoKimi K2.7 Code1.1T$0.95 / $4.00 /1M262K6h agoKimi K2 Thinking1.1T$0.60 / $2.50 /1M262K6h agoKimi K2 Thinking1.1T$0.60 / $2.50 /1M262K11m agoKimi K32.8T$3.00 / $15.00 /1M1M6h agoLing-2.6-1T1T$0.30 / $2.50 /1M262K6h agoLing-2.6-1T1T$0.075 / $0.63 /1M262K11m agoLing-2.6-flash107B$0.10 / $0.30 /1M262K6h agoLing-2.6-flash107B$0.010 / $0.030 /1M262K11m agoLlama 3.1 8B Instruct8B$0.020 / $0.050 /1M16K6h agoLlama 3.1 8B Instruct8B$0.020 / $0.050 /1M16K11m agoLlama 3.1 Euryale 70B v2.270.6B$1.45 / $1.45 /1M8K9m agoLlama 3.2 1B Instruct1.2B$0.020 / $0.020 /1M131K6h agoLlama 3.2 3B Instruct3.2B$0.030 / $0.050 /1M33K6h agoLlama 3.3 70B Instruct70.6B$0.14 / $0.40 /1M6K6h agoLlama 3.3 70B Instruct70.6B$0.14 / $0.40 /1M6K11m agoLlama 3 8B Lunaris8B$0.050 / $0.050 /1M8K9m agoLlama 4 Maverick402B$0.27 / $0.85 /1M1M11m agoLlama 4 Maverick402B$0.27 / $0.85 /1M1M6h agoLlama 4 Scout109B$0.18 / $0.59 /1M131K11m agoLlama 4 Scout109B$0.18 / $0.59 /1M131K6h agoMiMo-V2.5311B$0.17 / $0.34 /1M1M6h agoMiMo-V2.5311B$0.17 / $0.34 /1M1M9m agoMiMo-V2.5-Pro1T$0.48 / $0.96 /1M1M9m agoMiMo-V2.5-Pro1T$0.52 / $1.04 /1M1M6h agoMiniMax M1$0.55 / $2.20 /1M1M6h agoMiniMax M1$0.55 / $2.20 /1M1M11m agoMiniMax M2229B$0.30 / $1.20 /1M205K6h agoMiniMax M2229B$0.30 / $1.20 /1M205K11m agoMiniMax M2.1229B$0.30 / $1.20 /1M205K6h agoMiniMax M2.1229B$0.30 / $1.20 /1M205K11m agoMiniMax M2.5229B$0.60 / $2.40 /1M205K6h agoMiniMax M2.5229B$0.30 / $1.20 /1M205K11m agoMiniMax M2.7229B$0.27 / $1.08 /1M205K11m agoMiniMax M2.7229B$0.30 / $1.20 /1M205K6h agoMiniMax M3427B$0.30 / $1.20 /1M1M6h agoMiniMax M3427B$0.30 / $1.20 /1M1M11m agoMistral Nemo12.2B$0.040 / $0.17 /1M60K6h agoMistral Nemo12.2B$0.040 / $0.17 /1M60K3d agoMythoMax 13B13B$0.090 / $0.090 /1M4K6h agoNemotron 3 Nano 30B A3B31.6B$0.050 / $0.20 /1M262K11m agoNemotron 3 Nano 30B A3B31.6B$0.050 / $0.20 /1M262K6h agoQwen2.5 72B Instruct72.7B$0.38 / $0.40 /1M32K6h agoQwen2.5 72B Instruct72.7B$0.38 / $0.40 /1M32K10m agoQwen2.5 7B Instruct7.6B$0.070 / $0.070 /1M32K6h agoQwen2.5 VL 72B Instruct73.4B$0.80 / $0.80 /1M33K6h agoQwen3 235B A22B Instruct 2507235B$0.090 / $0.58 /1M131K6h agoQwen3 235B A22B Instruct 2507235B$0.090 / $0.58 /1M131K9m agoQwen3 235B A22B Thinking 2507235B$0.30 / $3.00 /1M131K6h agoQwen3 235B A22B Thinking 2507235B$0.30 / $3.00 /1M131K9m agoQwen3 30B A3B Instruct 250730.5B$0.090 / $0.45 /1M41K6h agoQwen3 32B32.8B$0.10 / $0.45 /1M41K6h agoQwen3.5-122B-A10B125B$0.40 / $3.20 /1M262K6h agoQwen3.5-122B-A10B125B$0.40 / $3.20 /1M262K9m agoQwen3.5-27B27.8B$0.30 / $2.40 /1M262K6h agoQwen3.5-27B27.8B$0.30 / $2.40 /1M262K9m agoQwen3.5-35B-A3B36B$0.25 / $2.00 /1M262K6h agoQwen3.5 397B A17B403B$0.60 / $3.60 /1M262K9m agoQwen3.5 397B A17B403B$0.60 / $3.60 /1M262K6h agoQwen3.6 27B27.8B$0.60 / $3.60 /1M262K6h agoQwen3.6 35B A3B36B$0.25 / $1.49 /1M262K6h agoQwen3.7 Max$1.25 / $3.75 /1M1M6h agoQwen3 8B8.2B$0.035 / $0.14 /1M128K6h agoQwen3 Coder 30B A3B Instruct30.5B$0.070 / $0.27 /1M160K6h agoQwen3 Coder 30B A3B Instruct30.5B$0.070 / $0.27 /1M160K9m agoQwen3 Coder 480B A35B480B$0.38 / $1.55 /1M262K6h agoQwen3 Coder 480B A35B480B$0.38 / $1.55 /1M262K9m agoQwen3 Coder Next79.7B$0.20 / $1.50 /1M262K9m agoQwen3 Coder Next79.7B$0.20 / $1.50 /1M262K6h agoQwen3 Max$2.11 / $8.45 /1M262K6h agoQwen3 Next 80B A3B Instruct81.3B$0.15 / $1.50 /1M131K9m agoQwen3 Next 80B A3B Instruct81.3B$0.15 / $1.50 /1M131K6h agoQwen3 Next 80B A3B Thinking81.3B$0.15 / $1.50 /1M131K6h agoQwen3 VL 235B A22B Instruct236B$0.30 / $1.50 /1M131K44h agoQwen3 VL 235B A22B Instruct236B$0.30 / $1.50 /1M131K6h agoQwen3 VL 235B A22B Thinking236B$0.98 / $3.95 /1M131K9m agoQwen3 VL 235B A22B Thinking236B$0.98 / $3.95 /1M131K6h agoQwen3 VL 30B A3B Instruct31.1B$0.20 / $0.70 /1M131K9m agoQwen3 VL 30B A3B Instruct31.1B$0.20 / $0.70 /1M131K6h agoQwen3 VL 30B A3B Thinking31.1B$0.20 / $1.00 /1M131K6h agoQwen3 VL 8B Instruct8.8B$0.080 / $0.50 /1M131K6h agoR1 0528685B$0.70 / $2.50 /1M164K11m agoR1 0528685B$0.70 / $2.50 /1M164K6h agoR1 Distill Llama 70B70.6B$0.80 / $0.80 /1M8K6h agoR1 Distill Llama 70B70.6B$0.80 / $0.80 /1M8K11m agoRing-2.6-1T$0.075 / $0.63 /1M262K11m agoRing-2.6-1T$0.30 / $2.50 /1M262K6h agoStep 3.7 Flash201B$0.20 / $1.15 /1M262K6h agoStep 3.7 Flash201B$0.20 / $1.15 /1M262K9m agoWizardLM-2 8x22B$0.62 / $0.62 /1M66K6h agoWizardLM-2 8x22B$0.62 / $0.62 /1M66K11m ago
Something wrong on this page? Tell us