$ diff qwen4-max k2-horizon
Qwen4-Max vs K2 Horizon
Values come from Qwen4-Max and K2 Horizon, where each is cited to its source. This page states no benchmark result and ranks neither model — it puts two published specifications next to each other. Where a lab has not published a figure, the row says so rather than guessing.
Alibaba QwenQwen4-Max
- context
- 1M
- weights
- —
- $/M in
- $1.5
- $/M out
- $4.5
- context
- 524K
- weights
- open
- $/M in
- —
- $/M out
- —
What actually differs
- Context window
- Qwen4-Max takes 1M against 524K — modestly more room in a single request.
- Recency
- Qwen4-Max shipped 77 days after K2 Horizon (2026-11-19 vs 2026-09-03).
Full spec
| Attribute | Qwen4-Max | K2 Horizon |
|---|---|---|
| Developer | Alibaba Qwen | Ifm |
| Released | 2026-11-19 | 2026-09-03 |
| Context window | 1,000,000 tokens | 524,288 (flagship 375B-A23B, native from midtraining onward) |
| Pricing | Singapore International: $1.5/M input · $4.5/M output, thinking and non-thinking | not recorded |
| License | not recorded | Apache 2.0 (weights and code) |
| Availability | Alibaba Cloud Model Studio API; model ID `qwen4-max` | Hugging Face (weights, code, checkpoints, training data where redistributable) |