Choose the right model for your workflow
Compare benchmarks, pricing, and capabilities across all supported models. All at a fraction of the cost of proprietary alternatives.
Model Benchmarks
DeepSeek
DeepSeek V3.2 Reasoner
128K
Complex reasoning at lowest cost
ToolsStreamingThinking
$0.28/M in$0.42/M out
Moonshot AI
Kimi K2.5
256K
Multi-step tool calling
ToolsStreamingThinkingVision
$0.6/M in$3/M out
Zhipu AI
GLM-5
200K
Best tool-call reliability
ToolsStreamingThinkingVision
$1/M in$3.2/M out
DeepSeek
DeepSeek V3.2
128K
Best price/performance
ToolsStreaming
$0.28/M in$0.42/M out
Alibaba
Qwen 3.5 Plus
256K
Strong all-rounder
ToolsStreamingThinkingVision
$0.4/M in$2.4/M out
Mistral
Devstral 2
256K
Development-focused
ToolsStreaming
$0.4/M in$2/M out
Zhipu AI
GLM-4.7
200K
Balanced performance
ToolsStreamingThinkingVision
$0.6/M in$2.2/M out
Mistral
Mistral Large 3
256K
European provider, strong code
ToolsStreaming
$0.5/M in$1.5/M out
Mistral
Codestral 25.08
256K
Code-specialized
ToolsStreaming
$0.3/M in$0.9/M out
Alibaba
Qwen Turbo
1M
Massive context, ultra-cheap
ToolsStreaming
$0.05/M in$0.2/M out
Mistral
Mistral Small 3.2
128K
Fast, cheap utility model
ToolsStreaming
$0.1/M in$0.3/M out
Zhipu AI
GLM-4 Flash
128K
Zero-cost experimentation
ToolsStreaming
FREE
Pricing Comparison
All prices in USD per million tokens.
| Model | Provider | Input $/M | Output $/M | Context | Thinking | Vision | Best For |
|---|---|---|---|---|---|---|---|
| DeepSeek V3.2 | DeepSeek | $0.28 | $0.42 | 128K | - | - | Best price/performance |
| DeepSeek V3.2 Reasoner | DeepSeek | $0.28 | $0.42 | 128K | Yes | - | Complex reasoning at lowest cost |
| Kimi K2.5 | Moonshot AI | $0.60 | $3.00 | 256K | Yes | Yes | Multi-step tool calling |
| GLM-5 | Zhipu AI | $1.00 | $3.20 | 200K | Yes | Yes | Best tool-call reliability |
| GLM-4.7 | Zhipu AI | $0.60 | $2.20 | 200K | Yes | Yes | Balanced performance |
| GLM-4 Flash | Zhipu AI | FREE | FREE | 128K | - | - | Zero-cost experimentation |
| Qwen 3.5 Plus | Alibaba | $0.40 | $2.40 | 256K | Yes | Yes | Strong all-rounder |
| Qwen Turbo | Alibaba | $0.05 | $0.20 | 1M | - | - | Massive context, ultra-cheap |
| Mistral Large 3 | Mistral | $0.50 | $1.50 | 256K | - | - | European provider, strong code |
| Codestral 25.08 | Mistral | $0.30 | $0.90 | 256K | - | - | Code-specialized |
| Devstral 2 | Mistral | $0.40 | $2.00 | 256K | - | - | Development-focused |
| Mistral Small 3.2 | Mistral | $0.10 | $0.30 | 128K | - | - | Fast, cheap utility model |
Estimate your costs
See how Ava's supported models compare to proprietary alternatives.
Cost Calculator
Estimate your monthly cost based on usage.
50
2,000
1,000
Estimated Monthly Cost
DeepSeek V3.2$1.47/mo
Kimi K2.5$6.30/mo
GLM-5$7.80/mo
GLM-4 FlashFREE
Qwen 3.5 Plus$4.80/mo
Qwen Turbo$0.45/mo
Mistral Large 3$3.75/mo
Codestral$2.25/mo
Mistral Small 3.2$0.75/mo
Claude Sonnet 4$31.50/mo
GPT-4o$22.50/mo