Skip to main content
BenchLM
Data
Recommendation
Ranking data as of

Best Tool Use & Function Calling Models in 2026

As of November 19, 2026, the top model among tool use & function calling models on the BenchLM leaderboard is Qwen3.7 Plus with a score of 72.

The table compares sourced tool-calling results for model configurations. It does not rank packaged agent products or establish reliability for your own API and MCP workflows.

Full Rankings (20 ranked)

1
Qwen3.7 Plus
Alibaba·Proprietary·1M

72

sourced avg

vs #2
2
Claude Opus 4.8
Anthropic·Proprietary·1M

70.6

sourced avg

vs #3
3
GLM-5.1
Z.AI·Open Weight·203K

70.1

sourced avg

vs #4
4
MiniMax M3
MiniMax·Open Weight·1M

70.1

sourced avg

vs #5
5
Gemini 3.5 Flash
Google·Proprietary·1M

69.8

sourced avg

vs #6
6
Qwen3.7 Max
Alibaba·Proprietary·1M

68.8

sourced avg

vs #7
7
GPT-5.5
OpenAI·Proprietary·1M

67.8

sourced avg

vs #8
8
GLM-5.2
Z.AI·Open Weight·1M

67

sourced avg

vs #9
9
GPT-5.4
OpenAI·Proprietary·1.05M

65.5

sourced avg

vs #10
10
Spark-X2.6-4B
SparkLLM·Open Weight·1M

64.6

sourced avg

vs #11
11
MiniMax M2.7
MiniMax·Open Weight·200K

63.9

sourced avg

vs #12
12
LLaDA2.2-flash
InclusionAI·Open Weight·128K

62.7

sourced avg

vs #13
13
Kimi K2.6
Moonshot AI·Open Weight·256K

60.5

sourced avg

vs #14
14
Qwen3.6-35B-A3B
Alibaba·Open Weight·262K

60.1

sourced avg

vs #15
15
Qwen3.6 Plus
Alibaba·Proprietary·1M

60

sourced avg

vs #16
16
LFM3-1.7B
LiquidAI·Open Weight·256K

58.9

sourced avg

vs #17
17
Claude Opus 4.5
Anthropic·Proprietary·200K

58.7

sourced avg

vs #18
18
GLM-5
Z.AI·Open Weight·200K

58.3

sourced avg

vs #19
19
Qwen3.5 397B
Alibaba·Open Weight·128K

57.7

sourced avg

vs #20
20
Kimi K2.5
Moonshot AI·Open Weight·256K

54.2

sourced avg

How to choose

Key Takeaways

The top model on this sourced reporting-family slice is Qwen3.7 Plus by Alibaba with an average of 72.

The best open-weight model is GLM-5.1 at position #3.

20 models are listed with sourced benchmark coverage in this reporting family.

Score in Context

What these scores mean

This ranking averages sourced tool-use benchmarks. It is narrower than the agentic category and focuses specifically on function calling and structured tool execution.

Known limitations

Tool-use benchmarks test specific function-calling patterns. Real-world tool use also depends on prompt engineering, retry logic, and error handling that benchmarks cannot capture.

About this ranking

Ranking data as of November 19, 2026

This reporting page focuses on structured output, tool routing, function calling, and MCP-style task completion. It is narrower than the general agentic leaderboard and is better aligned to developers choosing models for tool-heavy applications.

This page ranks models using only sourced tool-use benchmarks in the reporting family.

Qwen3.7 Plus leads this ranking with a score of 72, followed by Claude Opus 4.8 (70.6) and GLM-5.1 (70.1). The top three are separated by just a few points — any of them would perform well for this use case.

The best open-weight option is GLM-5.1 (ranked #3 with a score of 70.1). Open-weight models are highly competitive in this category — self-hosting is a viable alternative to proprietary APIs.

This ranking uses provisional overall weighted scores from the active scoring formula. For detailed model profiles, click any model name above. To compare two specific models head-to-head, use the "vs #" links.

Know when it’s worth switching models

The model to choose, the cheaper alternative, and the release we would wait on.

Read a sample issue

Join 7,000+ subscribers.

One email each week. Unsubscribe anytime.