o11y-bench 1.1 Pass@3

Name: o11y-bench 1.1 Pass@3 results
Creator: Village

Agents

Share of the 64 o11y-bench 1.1 tasks where at least one of three attempts passes.

Benchmark source: O11ybench

Top models (higher is better)

Model	Score	Source
Claude Haiku 5.5	100.0	o11y-bench 1.1 leaderboard: claude-haiku-5-5 (high) · Grafana Labs
Claude Sonnet 5.5	100.0	o11y-bench 1.1 leaderboard: claude-sonnet-5-5 (high) · Grafana Labs
GPT-6 Luna	100.0	o11y-bench 1.1 leaderboard: gpt-6-luna (high) · Grafana Labs
Claude Opus 5.5	98.4	o11y-bench 1.1 leaderboard: claude-opus-5-5 (high) · Grafana Labs
Gemini 4 Argon	98.4	o11y-bench 1.1 leaderboard: gemini-4-argon (low) · Grafana Labs
GLM-5.3	98.4	o11y-bench 1.1 leaderboard: accounts/fireworks/models/glm-5p3 (high) · Grafana Labs
GPT-6.1 Sol	96.9	o11y-bench 1.1 leaderboard: gpt-6.1-sol (low) · Grafana Labs
GPT-6 Astra	96.9	o11y-bench 1.1 leaderboard: gpt-6-astra (low) · Grafana Labs
Grok 4.7	96.9	o11y-bench 1.1 leaderboard: x-ai/grok-4.7 (low) · Grafana Labs
Kimi K3	96.9	o11y-bench 1.1 leaderboard: accounts/fireworks/models/kimi-k3 (high) · Grafana Labs
Mistral Large 4 Preview	96.9	o11y-bench 1.1 leaderboard: mistralai/mistral-large-4-0 (low) · Grafana Labs
Muse Spark Open	96.9	o11y-bench 1.1 leaderboard: accounts/fireworks/models/muse-spark-open (xhigh) · Grafana Labs
MiniMax M3.1 Flash	95.3	o11y-bench 1.1 leaderboard: minimax/minimax-m3.1-flash (low) · Grafana Labs
Qwen3.8 27B	95.3	o11y-bench 1.1 leaderboard: qwen/qwen3.8-27b (low) · Grafana Labs
DeepSeek V4.1 Flash 1017	93.8	o11y-bench 1.1 leaderboard: accounts/fireworks/models/deepseek-v4p1-flash-1017 (low) · Grafana Labs
GPT-5.4 Mini	92.2	o11y-bench 1.1 leaderboard: gpt-5.4-mini (low) · Grafana Labs
GPT-5.6 Luna	89.1	o11y-bench 1.1 leaderboard: gpt-5.6-luna (low) · Grafana Labs
Qwen3.6 27B	89.1	o11y-bench 1.1 leaderboard: qwen/qwen3.6-27b (low) · Grafana Labs
Gemma 4 31B IT	85.9	o11y-bench 1.1 leaderboard: google/gemma-4-31b-it (high) · Grafana Labs
Qwen3.6 35B A3B	81.3	o11y-bench 1.1 leaderboard: qwen/qwen3.6-35b-a3b (low) · Grafana Labs
Haiku 4.5	62.5	o11y-bench 1.1 leaderboard: claude-haiku-4-5 (off) · Grafana Labs
GPT-5.4 Nano	32.8	o11y-bench 1.1 leaderboard: gpt-5.4-nano (off) · Grafana Labs