Best LLMs for Tool Calling — Google
The best Google model for tool calling is Gemini 3.1 Pro Preview, completing 95.6% of Tau²-Bench's multi-turn tool-use tasks. Gemini 3.5 Flash (95.3%) and Gemini 2.5 Pro (54.1%) round out the top three.
- 1
gemini-3.1-pro-previewReasoningToolsJSON+246.5 intel · $2.00/M · 136 t/s95.6%τ²-Bench - 2gemini-3.5-flashReasoningToolsJSON+250.2 intel · $1.50/M · 275 t/s95.3%τ²-Bench
- 3gemini-2.5-proReasoningToolsJSON+225.8 intel · $1.25/M · 85 t/s54.1%τ²-Bench
- 4gemini-3.1-flash-liteReasoningToolsJSON+225.0 intel · $0.250/M · 337 t/s31.3%τ²-Bench
- 5gemini-3.1-flash-lite-previewReasoningToolsJSON+225.0 intel · $0.250/M · 337 t/s31.3%τ²-Bench
- 6gemini-3.6-flashReasoningToolsJSON+250.1 intel · $1.50/M · 135 t/s24.5%τ²-Bench
- 7gemini-2.5-flash-liteReasoningToolsJSON+26.9 intel · $0.100/M · 107 t/s19.0%τ²-Bench
- 8gemini-3.5-flash-liteReasoningToolsJSON+236.5 intel · $0.300/M · 160 t/s16.5%τ²-Bench
- 9gemini-2.5-flashReasoningToolsJSON+214.1 intel · $0.300/M · 150 t/s14.9%τ²-Bench
How this is ranked
AI models ranked by Tau²-Bench — multi-turn conversations where the model has to call the right tools, in the right order, against a real API to complete a customer task. This measures whether function calling actually works under pressure, which is a different question from whether a model supports the parameter at all.
Frequently asked
Which Google model is best at tool calling?
The best Google model for tool calling is Gemini 3.1 Pro Preview, completing 95.6% of Tau²-Bench's multi-turn tool-use tasks. Gemini 3.5 Flash (95.3%) and Gemini 2.5 Pro (54.1%) round out the top three.
What's a good alternative to Gemini 3.1 Pro Preview?
Gemini 3.5 Flash (95.3%) is the closest alternative on this metric, followed by Gemini 2.5 Pro (54.1%). See the full ranking above for the tradeoffs.
How many Google models are there?
modelgrep tracks 30 Google models with live benchmarks, speed, latency and per-provider pricing, led on intelligence by Gemini 3.5 Flash. 9 of them qualify for this ranking.