Chinese AI Model Rankings
Where the Chinese AI models ChinaAPI serves stand next to the international frontier, quoted from public leaderboards. Snapshot of 2026-08-26; roster last maintained 2026-08-26 (moves when a model or source is added). Every figure is reproduced from the board named in its column and links back to it. ChinaAPI runs no evaluation of its own and computes no composite score from these numbers. See how this list is built, the price of every model marked available, and the same data as JSON at /api/model-rankings.
Sources
- LMArena Text Arena — LMArena (arena.ai), Arena score (Elo), live board, retrieved 2026-08-26
- LiveBench — LiveBench (sponsored by Abacus.AI), Overall, LiveBench-2026-06-25, retrieved 2026-08-26
- Artificial Analysis Intelligence Index — Artificial Analysis, Intelligence Index, live board, retrieved 2026-08-26
- SuperCLUE 智能指数 — SuperCLUE, 总分 (overall), 2026-07 evaluation, published 2026-08-06, retrieved 2026-08-26
- Epoch Capabilities Index (ECI) — Epoch AI, General ECI, live board, retrieved 2026-08-26, CC BY 4.0, as labelled on the page (https://creativecommons.org/licenses/by/4.0/)
- Vals Index — Vals AI, Accuracy (GDP-weighted finance, coding and legal), V2 (released 2026-08-13), page states updated 8/19/2026, retrieved 2026-08-26
- Vendor's own evaluation card — the model's own vendor, varies; the benchmark is named in each row, self-reported, retrieved 2026-08-10
- LMArena Text-to-Image Arena — LMArena (arena.ai), Arena score (Elo), live board, retrieved 2026-08-26
- LMArena Image Edit Arena — LMArena (arena.ai), Arena score (Elo), live board, retrieved 2026-08-26
- LMArena Text-to-Video Arena — LMArena (arena.ai), Arena score (Elo), live board, retrieved 2026-08-26
- LMArena Image-to-Video Arena — LMArena (arena.ai), Arena score (Elo), live board, retrieved 2026-08-26
Text and multimodal models
Ordered by Epoch Capabilities Index (ECI); a model with no figure on that board follows the ones that have one, because absent from a board is not the same as last on it.
| # | Model | Vendor | Origin | Released | On ChinaAPI | LMArena Text Arena | LiveBench | Artificial Analysis Intelligence Index | SuperCLUE 智能指数 | Epoch Capabilities Index (ECI) | Vals Index | Vendor's own evaluation card | Notes |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Claude Fable 5 | Anthropic | International | 2026-06-09 | no | 1508 ±5 | 83.0 | 62 | — | 162 (160 - 166) | 66.04 ±1.03 | — | |
| 2 | Claude Opus 5 | Anthropic | International | 2026-07-24 | no | 1493 ±5 | 80.1 | 63 | — | 162 (159 - 165) | 67.21 ±0.98 | — | |
| 3 | GPT-5.6 Sol | OpenAI | International | 2026-07-09 | no | 1482 ±5 | 81.0 | 61 | — | 161 (159 - 165) | 63.71 ±1.06 | — | |
| 4 | GPT-5.5 | OpenAI | International | 2026-04-23 | no | 1482 ±4 | 80.2 | — | — | 159 (156 - 162) | 57.41 ±1.15 | — | |
| 5 | GPT-5.6 Terra | OpenAI | International | 2026-07-09 | no | 1465 ±5 | 77.9 | 57 | — | 159 (156 - 162) | 59.59 ±1.33 | — | |
| 6 | Gemini 3.7 Flash | International | 2026-08-13 | no | 1490 ±8 | 78.8 | 56 | — | 157 (155 - 160) | 59.31 ±1.06 | — | Released 2026-08-13. LMArena marks the row preliminary (5,718 votes as of 2026-08-26). | |
| 7 | Kimi K3 | Moonshot AI | China | 2026-07-16 | yes | 1489 ±6 | 79.2 | 60 | 70.68 | 157 (155 - 161) | 57.81 ±1.06 | — | |
| 8 | Grok 4.6 | xAI | International | 2026-08-12 | no | 1461 ±10 | 78.0 | 61 | — | 156 (154 - 158) | 59.17 ±1.21 | — | Released 2026-08-12. LMArena labels this vendor SpaceXAI and marks the row preliminary (3,473 votes as of 2026-08-26). |
| 9 | Qwen3.8-Max | Alibaba | China | 2026-07-19 | yes | 1481 ±7 | 78.5 | 58 | 71.48 | 156 (155 - 159) | 51.84 ±1.29 | — | |
| 10 | Gemini 3.1 Pro | International | 2026-02-19 | no | 1486 ±3 | 77.0 | 48 | — | 155 (153 - 157) | 41.90 ±1.17 | — | ||
| 11 | DeepSeek V4 Flash 0731 | DeepSeek | China | 2026-07-31 | yes | — | 74.2 | 52 | 65.60 | 154 (152 - 156) | 53.57 ±1.21 | — | The build the deepseek-v4-flash API serves. |
| 12 | Gemini 3.6 Flash | International | 2026-07-21 | yes | 1481 ±5 | 73.6 | 52 | — | 154 (153 - 156) | 55.35 ±1.09 | — | ||
| 13 | Grok 4.5 | xAI | International | 2026-07-08 | no | 1470 ±5 | 75.8 | 56 | — | 154 (152 - 156) | 51.53 ±1.32 | — | LMArena labels this vendor SpaceXAI. |
| 14 | Qwen3.7-Max | Alibaba | China | 2026-05-19 | yes | 1474 ±10 | 73.1 | — | — | 154 (152 - 156) | 44.77 ±1.08 | — | LMArena lists the preview build (qwen3.7-max-preview); the released model has no separate row. |
| 15 | GLM-5.2 | Zhipu AI (Z.ai) | China | 2026-06-16 | yes | 1470 ±5 | 73.2 | 53 | 63.27 | 152 (150 - 154) | 53.12 ±1.18 | — | |
| 16 | GLM-5.1 | Zhipu AI (Z.ai) | China | 2026-04-07 | yes | 1468 ±4 | — | — | — | 151 (149 - 152) | — | — | The Vals Index V2 (2026-08-13) no longer lists this model; the V1.2 figure it once carried was removed with that revision. |
| 17 | Kimi K2.6 | Moonshot AI | China | 2026-04-20 | yes | 1461 ±5 | 70.5 | — | — | 151 (149 - 153) | 43.47 ±1.17 | — | |
| 18 | Kimi K2.7 Code | Moonshot AI | China | 2026-06-12 | yes | — | 68.4 | 43 | — | 151 (149 - 152) | — | — | |
| 19 | DeepSeek V4 Pro | DeepSeek | China | 2026-04-24 | no | 1458 ±4 | — | 45 | — | 149 (148 - 151) | 42.89 ±1.19 | — | The pre-0813 build; the deepseek-v4-pro API moved to DeepSeek-V4-Pro-0813 on 2026-08-12. LiveBench dropped this row in favour of the dated builds between the 2026-08-19 and 2026-08-26 snapshots, so its 71.6 is no longer quoted. |
| 20 | Qwen3.6-Plus | Alibaba | China | 2026-04-01 | yes | — | 68.9 | — | — | 149 (147 - 151) | 31.98 ±0.98 | — | |
| 21 | MiniMax-M3 | MiniMax | China | 2026-06-01 | yes | — | 67.3 | 45 | 56.90 | 148 (143 - 150) | 42.72 ±1.20 | — | |
| 22 | Qwen3.7-Plus | Alibaba | China | 2026-06-02 | yes | 1457 ±5 | — | 39 | — | 148 (146 - 151) | 38.65 ±1.17 | — | |
| 23 | GLM-5 | Zhipu AI (Z.ai) | China | 2026-02-11 | yes | 1457 ±4 | — | — | — | 146 (144 - 148) | — | — | |
| 24 | MiniMax-M2.7 | MiniMax | China | 2026-03-18 | yes | — | — | — | — | 146 (139 - 148) | 24.56 ±0.93 | — | |
| 25 | Qwen3.6-Flash | Alibaba | China | 2026-04-26 | yes | — | — | — | — | 144 (141 - 146) | — | — | |
| 26 | DeepSeek V4 Pro 0813 | DeepSeek | China | 2026-08-13 | yes | 1459 ±10 | 77.4 | 53 | 70.10 | — | 52.37 ±1.14 | — | The build the deepseek-v4-pro API has served since 2026-08-12. LMArena's 0813 row now carries a rank and 3,621 human votes rather than the unranked AutoEval preview it was on 2026-08-18, so it is quoted here; Epoch still scores only the undated DeepSeek-V4-Pro and that figure is not reused for this build. |
| 27 | MiMo-V2.5-Pro | Xiaomi | China | 2026-04-22 | yes | 1468 ±4 | — | 43 | 53.01 | — | 40.97 ±1.38 | — | |
| 28 | Muse Spark 1.2 | Meta | International | 2026-08-05 | no | 1498 ±10 | 78.0 | 57 | — | — | 57.05 ±1.12 | — | |
| 29 | GLM-5.3 | Zhipu AI (Z.ai) | China | 2026-08-14 | yes | 1487 ±10 | 76.1 | 60 | — | — | — | — | Released 2026-08-14; served here via Zhipu's official channels since 2026-08-18. Independent boards picked it up between the 2026-08-19 and 2026-08-26 snapshots, so the vendor's own DeepSWE figure it carried before has been replaced by their readings; the LMArena row stands on 3,751 votes. Epoch does not score it, which is why it sits after the models this category's primary board covers. |
| 30 | Hunyuan Hy3 | Tencent | China | 2026-07-06 | yes | 1457 ±8 | — | 42 | 62.13 | — | — | — | |
| 31 | DeepSeek V4 Flash Vision Exp | DeepSeek | China | 2026-08-21 | yes | — | 76.8 | 51 | — | — | — | — | DeepSeek's experimental vision tier, announced in the vendor's change log on 2026-08-21 as model='deepseek-v4-flash-vision-exp'. LiveBench names its row with the vendor's -Exp suffix and Artificial Analysis drops it, but DeepSeek publishes only one vision tier, so both rows are this model. LMArena, Vals, SuperCLUE and Epoch carry no row for it, and the sibling flash figures are not reused; with no Epoch figure it sits after the models this category's primary board covers. |
| 32 | LongCat-2.0 | Meituan | China | 2026-06-30 | yes | — | — | 34 | 54.22 | — | — | — | |
| 33 | MiMo-V2.5 | Xiaomi | China | 2026-04-22 | yes | — | — | 38 | — | — | 39.91 ±1.31 | — | |
| 34 | Step 3.7 Flash | StepFun | China | 2026-05-29 | yes | — | — | 31 | 48.16 | — | — | — | |
| 35 | DeepSeek V4 Flash | DeepSeek | China | — | no | 1436 ±4 | — | — | — | — | — | — | The pre-0731 build. LMArena still lists it alongside 0731; LiveBench dropped its row between the 2026-08-19 and 2026-08-26 snapshots, so its 65.5 is no longer quoted. |
| 36 | Doubao-Seed-2.1-pro | ByteDance | China | 2026-06-28 | yes | — | — | — | 65.94 | — | — | Terminal Bench 2.1 71.0 | |
| 37 | GLM-5V-Turbo | Zhipu AI (Z.ai) | China | — | yes | 1433 ±7 | — | — | — | — | — | — | Vision-language model. LMArena's text board now carries a glm-5v-turbo row (9,380 votes); Epoch does not score it, so it sits after the models the primary board covers. |
| 38 | Agnes 2.0 Flash | Agnes AI | China | — | yes | — | — | — | — | — | — | Claw-Eval Pass^3 60.9% | No independent board in this snapshot covers it; the figure shown is the vendor's own. |
| 39 | Agnes 2.5 Flash | Agnes AI | China | — | yes | — | — | — | — | — | — | — | Artificial Analysis carries a different build of this family (Agnes 2.5 Pro Alpha, 40) and the vendor publishes its own figures only as a chart image, so there is no number here that can be quoted. |
| 40 | Doubao-Seed-2.1-turbo | ByteDance | China | 2026-06-28 | yes | — | — | — | — | — | — | Terminal Bench 2.1 67.6 | No independent board in this snapshot covers it; the figure shown is the vendor's own. The vendor writes the name without the 260628 snapshot suffix we serve it under. |
| 41 | GLM-5-Turbo | Zhipu AI (Z.ai) | China | 2026-03-16 | yes | — | — | — | — | — | — | — | No independent board in this snapshot carries a row for it. |
| 42 | Step 3.5 Flash | StepFun | China | 2026-02-02 | yes | — | — | — | — | — | — | SWE-bench Verified 74.4 | No independent board in this snapshot covers it; the figure shown is the vendor's own. |
Image generation models
Ordered by LMArena Text-to-Image Arena; a model with no figure on that board follows the ones that have one, because absent from a board is not the same as last on it.
| # | Model | Vendor | Origin | Released | On ChinaAPI | LMArena Text-to-Image Arena | LMArena Image Edit Arena | Notes |
|---|---|---|---|---|---|---|---|---|
| 1 | GPT Image 2 | OpenAI | International | 2026-04-21 | no | 1382 ±4 | 1462 ±4 | |
| 2 | Gemini 3.1 Flash Image | International | — | no | 1263 ±5 | 1387 ±4 | ||
| 3 | Doubao Seedream 5.0 | ByteDance | China | 2026-01-28 | yes | 1258 ±5 | 1395 ±4 | Scores are the arena's seedream-5.0-pro rows; the arena also carries a seedream-5.0-lite build (1136 / 1293). |
| 4 | FLUX.2 Pro | Black Forest Labs | International | 2025-11-25 | no | 1155 ±3 | 1245 ±3 | |
| 5 | Doubao Seedream 4.5 | ByteDance | China | 2025-11-28 | yes | 1147 ±3 | 1301 ±2 | |
| 6 | Wan 2.7 Image Pro | Alibaba | China | 2026-04-01 | yes | 1103 ±5 | 1302 ±4 | |
| 7 | Wan 2.7 Image | Alibaba | China | 2026-04-01 | yes | 1100 ±5 | 1301 ±4 | |
| 8 | Agnes Image 2.0 Flash | Agnes AI | China | — | yes | — | — | No row on the LMArena image boards at the time of this snapshot. |
| 9 | Agnes Image 2.1 Flash | Agnes AI | China | — | yes | — | — | No row on the LMArena image boards at the time of this snapshot. |
| 10 | MiniMax image-01 | MiniMax | China | — | yes | — | — | No row on the LMArena image boards at the time of this snapshot. |
| 11 | Step 2X Large | StepFun | China | — | yes | — | — | No row on the LMArena image boards at the time of this snapshot; the arena carries an older StepFun build (step1x-edit, 998) that is a different model. |
| 12 | Step Image Edit 2 | StepFun | China | — | yes | — | — | No row on the LMArena image boards at the time of this snapshot; the arena carries an older StepFun build (step1x-edit, 998) that is a different model. |
Video generation models
Ordered by LMArena Text-to-Video Arena; a model with no figure on that board follows the ones that have one, because absent from a board is not the same as last on it.
| # | Model | Vendor | Origin | Released | On ChinaAPI | LMArena Text-to-Video Arena | LMArena Image-to-Video Arena | Notes |
|---|---|---|---|---|---|---|---|---|
| 1 | Gemini Omni Flash (video) | International | — | no | 1512 ±11 | 1462 ±6 | ||
| 2 | Doubao Seedance 2.0 | ByteDance | China | 2026-01-28 | yes | 1482 ±10 | 1478 ±8 | The arena entry is the 720p Dreamina build of Seedance 2.0. |
| 3 | Doubao Seedance 2.5 | ByteDance | China | 2026-06-28 | yes | 1477 ±19 | 1484 ±12 | The LMArena video boards picked up dreamina-seedance-2.5-720p between the 2026-08-19 and 2026-08-26 snapshots. Both rows are still thin — 1,337 text-to-video and 2,912 image-to-video votes — and their intervals are correspondingly wide. |
| 4 | MiniMax H3 | MiniMax | China | 2026-07-31 | yes | 1453 ±13 | 1489 ±7 | |
| 5 | Sora 2 Pro | OpenAI | International | 2025-09-30 | no | 1364 ±7 | — | |
| 6 | Veo 3.1 | International | 2025-10-15 | no | 1364 ±14 | 1397 ±11 | ||
| 7 | Wan 2.7 T2V | Alibaba | China | 2026-04-03 | yes | 1344 ±9 | — | |
| 8 | MiniMax Hailuo 2.3 | MiniMax | China | 2025-10-28 | yes | 1204 ±7 | 1261 ±5 | |
| 9 | MiniMax Hailuo 02 | MiniMax | China | — | yes | 1198 ±12 | 1228 ±11 | Scores are the arena's hailuo-02-pro rows; the arena also carries standard and fast builds of the same family. |
| 10 | Kling v3 | Kuaishou | China | 2026-02-05 | yes | — | 1356 ±6 | The arena entry is the pro build; the text-to-video board has no v3 row at the time of this snapshot. |
| 11 | Wan 2.7 I2V | Alibaba | China | 2026-04-03 | yes | — | 1427 ±5 | |
| 12 | HappyHorse 1.1 I2V | Alibaba | China | — | yes | — | — | The LMArena video boards carry the previous generation (happyhorse-1.0, 1428 text-to-video / 1442 image-to-video); that score is not this model's and is deliberately not reused here. |
| 13 | HappyHorse 1.1 T2V | Alibaba | China | — | yes | — | — | The LMArena video boards carry the previous generation (happyhorse-1.0, 1428 text-to-video / 1442 image-to-video); that score is not this model's and is deliberately not reused here. |
| 14 | Kling 3.0 Turbo | Kuaishou | China | — | yes | — | — | No row on the LMArena video boards at the time of this snapshot. |
| 15 | Kling v3 Omni | Kuaishou | China | 2026-02-05 | yes | — | — | No row on the LMArena video boards at the time of this snapshot. |
| 16 | MiniMax Hailuo 2.3 Fast | MiniMax | China | 2025-10-28 | yes | — | — | No row on the LMArena video boards at the time of this snapshot. |
Real usage on ChinaAPI
Which models the calls ChinaAPI actually served went to, as positions and shares of this platform's own traffic. It is a measurement of one gateway, not of the Chinese AI market, and it is not comparable with the benchmark results above. No call count, token count or revenue figure is published here or anywhere else on this site: every number below is a proportion. Recomputed hourly. A model with too little traffic in a window to rank is left unnamed and counted under “other”, as is anything outside the published catalogue. The same board as JSON is at /api/model-rankings/usage.
Share of calls served, the last 7 days
| # | Model | Vendor | Share of calls | Trend vs the previous 7 days |
|---|---|---|---|---|
| 1 | deepseek-v4-flash | DeepSeek | 10.4% | down |
| 2 | agnes-2.5-flash | Agnes AI | 6.9% | down |
| 3 | deepseek-v4-pro | DeepSeek | 4.4% | down |
| 4 | gpt-5.6-sol | OpenAI | 4.2% | flat |
| 5 | gemini-3.6-flash | 3.0% | down | |
| 6 | kimi-k2.7-code | Moonshot | 2.9% | up |
| 7 | kimi-k3 | Moonshot | 2.7% | up |
| 8 | gpt-5.6-luna | OpenAI | 2.3% | down |
| 9 | claude-opus-4-8 | Anthropic | 2.1% | down |
| 10 | glm-5.3-flash | 智谱 | 2.0% | up |
| 11 | mimo-v2.5 | Xiaomi | 1.8% | up |
| 12 | deepseek-v4-flash-vision-exp | DeepSeek | 1.7% | up |
| 13 | gpt-5.6-terra | OpenAI | 1.7% | down |
| 14 | agnes-image-2.1-flash | Agnes AI | 1.7% | up |
| 15 | gpt-5.5 | OpenAI | 1.6% | down |
| 16 | claude-sonnet-5 | Anthropic | 1.5% | down |
| 17 | claude-opus-5 | Anthropic | 1.5% | down |
| 18 | glm-5.2 | 智谱 | 1.3% | flat |
| 19 | MiniMax-M2.7 | MiniMax | 1.3% | flat |
| 20 | claude-opus-4-7 | Anthropic | 1.3% | flat |
| All other models | 43.6% |
How those calls arrived: 9.2% over agent protocols (Claude Messages and the OpenAI Responses API), 86.2% over chat protocols (OpenAI chat completions and Gemini generateContent), 4.6% over everything else (audio, images, embeddings, rerank and asynchronous tasks).
Share of calls served, the last 30 days
| # | Model | Vendor | Share of calls | Trend vs the previous 30 days |
|---|---|---|---|---|
| 1 | deepseek-v4-flash | DeepSeek | 21.6% | down |
| 2 | deepseek-v4-pro | DeepSeek | 14.5% | down |
| 3 | mimo-v2.5 | Xiaomi | 6.3% | down |
| 4 | agnes-2.5-flash | Agnes AI | 3.9% | up |
| 5 | claude-sonnet-5 | Anthropic | 3.2% | up |
| 6 | gpt-5.6-sol | OpenAI | 2.9% | up |
| 7 | kimi-k3 | Moonshot | 2.5% | down |
| 8 | gemini-3.6-flash | 2.3% | up | |
| 9 | gpt-5.6-luna | OpenAI | 2.3% | up |
| 10 | claude-opus-4-8 | Anthropic | 1.9% | up |
| 11 | gpt-5.6-terra | OpenAI | 1.9% | up |
| 12 | claude-opus-5 | Anthropic | 1.7% | up |
| 13 | kimi-k2.7-code | Moonshot | 1.6% | up |
| 14 | gpt-5.5 | OpenAI | 1.6% | up |
| 15 | glm-5.2 | 智谱 | 1.5% | down |
| 16 | gemini-3.1-flash-lite | 1.2% | up | |
| 17 | gpt-4.1-mini | OpenAI | 1.1% | up |
| 18 | MiniMax-M2.7 | MiniMax | 0.9% | down |
| 19 | claude-fable-5 | Anthropic | 0.9% | flat |
| 20 | glm-5 | 智谱 | 0.9% | flat |
| All other models | 25.2% |
How those calls arrived: 6.3% over agent protocols (Claude Messages and the OpenAI Responses API), 67.3% over chat protocols (OpenAI chat completions and Gemini generateContent), 26.4% over everything else (audio, images, embeddings, rerank and asynchronous tasks).