Alternatives to Qwen2.5 VL 72B Instruct
These models are alternatives by measurable coding evidence: they share the current open/open-weight classificationand are closest to Qwen2.5 VL 72B Instruct's SWE-bench score. Similarity here does not mean identical capabilities, licence terms, or output quality.
The target model
SWE-bench
73.5%
Cheapest input
$0.25
Cheapest output
$0.75
Hosts
5
| Alternative | SWE-bench | Input $/MTok | Output $/MTok | Hosts | Context | Tradeoff |
|---|---|---|---|---|---|---|
| Kimi K2.5 | 73.8% | $0.40 | $1.90 | 30 | 262k | higher input price |
| Kimi K2.6 | 76.7% | $0.55 | $2.52 | 48 | 256k | higher input price |
| Qwen3.6 Max Preview | 76.7% | $1.03 | $6.16 | 2 | 262k | higher input price |
| Qwen3.6 27B | 70.0% | $0.29 | $2.00 | 21 | 262k | higher input price |
| Qwen3.7 Max | 77.3% | $1.48 | $4.42 | 2 | 1.0M | higher input price |
| DeepSeek V4 Pro 0423 | 77.6% | $0.79 | $1.74 | 41 | 1.0M | higher input price |
| Nemotron 3 Ultra | 69.0% | $0.50 | $2.20 | 10 | 262k | higher input price |
| Kimi K2.7 Code | 78.2% | $0.67 | $3.40 | 35 | 262k | higher input price |
| Devstral 2 2512 | 62.8% | $0.44 | $2.20 | 2 | 262k | higher input price |
| Qwen3.8 Max | 85.6% | $2.00 | $6.00 | 2 | 1.0M | higher coding score, higher input price |
| Qwen3.8 27B | 86.0% | $0.35 | $2.55 | 20 | 262k | higher coding score, higher input price |
| Kimi K2 0711 | 60.2% | $0.57 | $2.30 | 3 | 131k | higher input price |
How alternatives are selected
Alternatives must have a benchmark score and the same current open/closed classification as the target. They are sorted by score distance, then by price/performance. The table compares hosted API economics; it does not establish licence compatibility or self-hosting rights.
Check the model licence, provider Terms, context limits, tool support, and output quality before switching a production workload.