Composite "best for X" tables ranked from live catalog data, plus per-benchmark top 10s at the bottom. Only official and auditable third-party benchmark scores — no in-house evals.
Lowest blended price / 1M tokens for models tagged chat or multilingual. Picked the cheapest hosting per model.
| # | Model | Cheapest / 1M | Via | Context |
|---|---|---|---|---|
| 1 | Doubao Lite 32K ByteDance | $0.05 | ByteDance Doubao (Volcengine) | 33K |
| 2 | GLM-4-Air Zhipu AI | $0.07 | Zhipu AI | 131K |
| 3 | Yi Lightning 01.AI | $0.14 | 01.AI | 16K |
| 4 | Step 3.5 Flash StepFun (阶跃星辰) · open-weight | $0.15 | StepFun | 262K |
| 5 | Doubao Pro 32K ByteDance | $0.15 | ByteDance Doubao (Volcengine) | 33K |
| 6 | DeepSeek V3.1 DeepSeek · open-weight | $0.30 | DeepSeek | 33K |
| 7 | DeepSeek V4 Flash DeepSeek · open-weight | $0.33 | DeepSeek | 1049K |
| 8 | DeepSeek V3 DeepSeek AI · open-weight | $0.48 | DeepSeek | 66K |
| 9 | MiniMax M2.7 MiniMax · open-weight | $0.52 | MiniMax | 197K |
| 10 | Qwen3.5-122B-A10B Alibaba Cloud · open-weight | $0.71 | Alibaba Cloud DashScope | 262K |
Lowest blended price for reasoning / chain-of-thought / tool-use models. Same pricing methodology as the chat board.
| # | Model | Cheapest / 1M | Via | Context |
|---|---|---|---|---|
| 1 | GLM-4-Air Zhipu AI | $0.07 | Zhipu AI | 131K |
| 2 | Yi Lightning 01.AI | $0.14 | 01.AI | 16K |
| 3 | Step 3.5 Flash StepFun (阶跃星辰) · open-weight | $0.15 | StepFun | 262K |
| 4 | Doubao Pro 32K ByteDance | $0.15 | ByteDance Doubao (Volcengine) | 33K |
| 5 | DeepSeek V3.1 DeepSeek · open-weight | $0.30 | DeepSeek | 33K |
| 6 | DeepSeek V3 DeepSeek AI · open-weight | $0.48 | DeepSeek | 66K |
| 7 | Qwen3.5-122B-A10B Alibaba Cloud · open-weight | $0.71 | Alibaba Cloud DashScope | 262K |
| 8 | Qwen3.5 397B A17B Alibaba Cloud · open-weight | $0.88 | Alibaba Cloud DashScope | 262K |
| 9 | Llama 3.3 70B Instruct Meta · open-weight | $0.88 | Together.ai | 131K |
| 10 | Qwen3 72B Instruct Alibaba · open-weight | $0.90 | Together.ai | 131K |
Chat / reasoning / multimodal models ranked by context window. For document Q&A, whole-repo code review, or long transcripts. Embedding / image-gen / video types excluded — their 'context' is a different quantity.
| # | Model | Cheapest / 1M | Via | Context |
|---|---|---|---|---|
| 1 | DeepSeek V4 Flash 0731 DeepSeek · open-weight | $0.17 | DeepSeek | 1311K |
| 2 | MiMo-V2.5-Pro Xiaomi · open-weight | $0.54 | Xiaomi MiMo | 1050K |
| 3 | MiMo-V2.5 Xiaomi · open-weight | $0.17 | Xiaomi MiMo | 1050K |
| 4 | DeepSeek V4 Flash DeepSeek · open-weight | $0.33 | DeepSeek | 1049K |
| 5 | DeepSeek V4 Pro DeepSeek · open-weight | $0.99 | DeepSeek | 1049K |
| 6 | MiniMax M3 MiniMax · open-weight | $0.53 | MiniMax | 1049K |
| 7 | Kimi K3 Moonshot AI · open-weight | $6.00 | Moonshot AI | 1049K |
| 8 | GLM-5.2 Zhipu AI · open-weight | $2.15 | Zhipu AI | 1049K |
| 9 | GLM 5.3 Zhipu AI | $2.15 | Zhipu AI | 1049K |
| 10 | Qwen3.8 2.4T A95B Alibaba Cloud · open-weight | $3.00 | Alibaba Cloud DashScope | 1049K |
Models that ship weights you can self-host. Ordered by context window — biggest first.
| # | Model | Cheapest / 1M | Via | Context |
|---|---|---|---|---|
| 1 | DeepSeek V4 Flash 0731 DeepSeek · open-weight | $0.17 | DeepSeek | 1311K |
| 2 | MiMo-V2.5-Pro Xiaomi · open-weight | $0.54 | Xiaomi MiMo | 1050K |
| 3 | MiMo-V2.5 Xiaomi · open-weight | $0.17 | Xiaomi MiMo | 1050K |
| 4 | DeepSeek V4 Flash DeepSeek · open-weight | $0.33 | DeepSeek | 1049K |
| 5 | DeepSeek V4 Pro DeepSeek · open-weight | $0.99 | DeepSeek | 1049K |
| 6 | MiniMax M3 MiniMax · open-weight | $0.53 | MiniMax | 1049K |
| 7 | Kimi K3 Moonshot AI · open-weight | $6.00 | Moonshot AI | 1049K |
| 8 | GLM-5.2 Zhipu AI · open-weight | $2.15 | Zhipu AI | 1049K |
| 9 | Qwen3.8 2.4T A95B Alibaba Cloud · open-weight | $3.00 | Alibaba Cloud DashScope | 1049K |
| 10 | DeepSeek V4 Pro 0813 DeepSeek · open-weight | $1.78 | DeepSeek | 1049K |
Filter: `code` capability. Ranked by top HumanEval / LiveCodeBench / SWE-bench score when available — otherwise the model is hidden to avoid a misleading placeholder.
| # | Model | Cheapest / 1M | Via | Context | Score |
|---|---|---|---|---|---|
| 1 | R1 DeepSeek · open-weight | $1.15 | DeepSeek | 64K | 89.1% HumanEval |
| 2 | Llama 3.3 70B Instruct Meta · open-weight | $0.88 | Together.ai | 131K | 88.4% HumanEval |
| 3 | DeepSeek V3 DeepSeek AI · open-weight | $0.48 | DeepSeek | 66K | 82.5% HumanEval |
Models whose every published hosting sits on a Provider without an overseas node. For CN-mainland users who don't want cross-border egress.
| # | Model | Cheapest / 1M | Via | Context |
|---|---|---|---|---|
| 1 | GLM-4 Plus Zhipu AI | $0.00 | Zhipu AI | 128K |
| 2 | GLM 4.5 Air (free) Zhipu AI · open-weight | $0.00 | Zhipu AI | 131K |
| 3 | Kimi K2.6 (free) Moonshot AI · open-weight | $0.00 | Moonshot AI | 262K |
| 4 | Doubao Lite 32K ByteDance | $0.05 | ByteDance Doubao (Volcengine) | 33K |
| 5 | GLM-4-Air Zhipu AI | $0.07 | Zhipu AI | 131K |
| 6 | Hy-MT2-1.8B Tencent · open-weight | $0.08 | Tencent Hunyuan | 8K |
| 7 | GLM 4 32B Zhipu AI · open-weight | $0.10 | Zhipu AI | 128K |
| 8 | Hy3 preview Tencent · open-weight | $0.11 | Tencent Hunyuan | 262K |
| 9 | UI-TARS 7B ByteDance · open-weight | $0.13 | ByteDance Doubao (Volcengine) | 128K |
| 10 | Hy-MT2-30B-A3B Tencent · open-weight | $0.13 | Tencent Hunyuan | 8K |
Existing published benchmark scores — sourced from papers or auditable third-party boards. Click any model to see its full profile.
| # | Model | Score | Conditions | Source |
|---|---|---|---|---|
| 1 | Hunyuan-Large Tencent | 90.00 | instruct model | official |
| 2 | Hunyuan-Large Tencent | 90.00 | instruct | official |
| 3 | MiniMax-Text-01 MiniMax | 86.90 | — | official |
| 4 | Hunyuan-Large Tencent | 71.40 | pre-trained model | official |
| 5 | Hunyuan-Large Tencent | 71.40 | pretrained | official |
| 6 | R1 DeepSeek | 0.89 | 0-shot pass@1 | official |
| 7 | Llama 3.3 70B Instruct Meta | 0.88 | 0-shot pass@1 | official |
| 8 | DeepSeek V3 DeepSeek AI | 0.82 | 0-shot pass@1 | official |
| 9 | Qwen3 72B Instruct Alibaba | 0.82 | 0-shot pass@1 | official |
| # | Model | Score | Conditions | Source |
|---|---|---|---|---|
| 1 | Hunyuan-Large Tencent | 89.90 | instruct model | official |
| 2 | Hunyuan-Large Tencent | 89.90 | instruct | official |
| 3 | MiniMax-Text-01 MiniMax | 88.50 | — | official |
| 4 | Hunyuan-Large Tencent | 88.40 | pre-trained model | official |
| 5 | Hunyuan-Large Tencent | 88.40 | pretrained | official |
| 6 | R1 DeepSeek | 0.90 | 5-shot | official |
| 7 | DeepSeek V3 DeepSeek AI | 0.89 | 5-shot | official |
| 8 | Llama 3.3 70B Instruct Meta | 0.87 | 5-shot | official |
| 9 | Qwen3 72B Instruct Alibaba | 0.86 | 5-shot | official |
Fast Pivot — Access hundreds of AI models from a single API — competitive pricing, reliable uptime, enterprise-grade infrastructure.