← the full leaderboard · open weights
The best open-weights AI models, ranked.
Kimi K3 is the highest-ranked open-weights model on the ArtificialWatch Bench Index (#11 of 31). Open-weights models ranked against every model on the index, with license, size and Hugging Face repo. Updated daily.
Scores as published by each board; table composed September 28, 2026.
Get the next launch alert, free →
Open-weights leaderboard
| # | Model | AW index | Coding index | Size | License | Weights | API price / 1M |
|---|---|---|---|---|---|---|---|
| 1 | Kimi K3 · Moonshot | 54.1 · #11 of 31 | 80 · #7 | 2.8T | custom (kimi-k3) | Hugging Face | $3 / $15 |
| 2 | DeepSeek V4.1 Flash · DeepSeek | 49.2 · #13 of 31 | 72.4 · #8 | 763B | MIT | Hugging Face | — |
| 3 | GLM-5.3 · Zhipu AI | 42.4 · #17 of 31 | 48.3 · #16 | 753B | custom (glm-5.3) | Hugging Face | $1.40 / $4.40 |
| 4 | GLM-5.3 Flash · Zhipu AI | 36 · #20 of 31 | 51.7 · #14 | 321B | MIT | Hugging Face | $0.15 / $0.50 |
| 5 | DeepSeek V4 Pro · DeepSeek | 28.8 · #23 of 31 | 40 · #19 | 1.65T | MIT | Hugging Face | $0.435 / $0.87 |
| 6 | GLM-5.2 · Zhipu AI | 23.3 · #25 of 31 | 26.7 · #23 | 753B | MIT | Hugging Face | $1.40 / $4.40 |
| 7 | MiniMax M3 · MiniMax | 6.5 · #29 of 31 | 6.9 · #30 | 427B | custom (minimax-community) | Hugging Face | $0.30 / $1.20 |
| 8 | Inkling · Thinking Machines | 0 · #31 of 31 | 3.4 · #31 | 952B | Apache 2.0 | Hugging Face | — |
AW index: the median of each model's percentile across the public leaderboards that score it, ranked against all 31 models on the full leaderboard. Coding index: the same rule over three coding boards (the coding leaderboard). Size: total parameters as Hugging Face counts the repo's weights. License: as the repo lists it. Prices are input / output per 1M tokens on the maker's own API.
Best open-source LLM for coding
- Kimi K3 · coding index 80 · #7 of 31 on the coding leaderboard
- DeepSeek V4.1 Flash · coding index 72.4 · #8 of 31 on the coding leaderboard
- GLM-5.3 Flash · coding index 51.7 · #14 of 31 on the coding leaderboard
- GLM-5.3 · coding index 48.3 · #16 of 31 on the coding leaderboard
- DeepSeek V4 Pro · coding index 40 · #19 of 31 on the coding leaderboard
- GLM-5.2 · coding index 26.7 · #23 of 31 on the coding leaderboard
- MiniMax M3 · coding index 6.9 · #30 of 31 on the coding leaderboard
- Inkling · coding index 3.4 · #31 of 31 on the coding leaderboard
Ordered by coding rank. The full AI coding leaderboard →
Quick answers
What is the best open-weights (open-source) AI model right now?
Kimi K3 from Moonshot: the highest-ranked model with downloadable weights on the ArtificialWatch Bench Index, at 54.1 (#11 of 31 models, open and closed). Next are DeepSeek V4.1 Flash (49.2) and GLM-5.3 (42.4). Its weights are on Hugging Face as moonshotai/Kimi-K3, license custom (kimi-k3).
What is the best open-weights model for coding?
Kimi K3, at 80 on the ArtificialWatch coding index (#7 of 31 models, open and closed).
How far behind closed models are open-weights models?
The best open-weights model, Kimi K3, ranks #11 of 31 on the Bench Index at 54.1; 10 models rank above it, led by Claude Opus 5.5 at 100. The index is the median percentile across public leaderboards (0-100), so the gap is a gap in rank on those boards, not a measure of how much more a model can do.
Is open-weights the same as open source?
Not quite. Open-weights means the trained weights can be downloaded and run on your own hardware. The Open Source Initiative's definition also covers the license terms and detailed information about the training data. Several models here ship under their maker's own license rather than a standard one like MIT or Apache 2.0, and each row names the license its Hugging Face repo lists.
Which models count as open-weights here?
Models whose weights are public on Hugging Face and that at least 2 public leaderboards score: Kimi K3, DeepSeek V4.1 Flash, GLM-5.3, GLM-5.3 Flash, DeepSeek V4 Pro, GLM-5.2, MiniMax M3 and Inkling. The repos were last checked September 28, 2026.
More rankings
The full LLM leaderboard · The AI coding leaderboard · Every model's API price · Head-to-head comparisons
Hear the minute the next one drops. Free.
A free Chrome push the moment a new model answers on a public API, an email about 15 minutes later. Texts or a phone call on paid plans.
- Free forever
- No card
- Unsubscribe in one click