Skip to content

Benchmark

SWE-Bench Pro

Agentic benchmarks measure how well models carry out tasks that take several steps. Scores are percentages. Higher is better.

Models and systems measured
93
Scores recorded
188
Sources
4
IQ weight
3.48

Score distribution

57.3%
Mean
57.7%
Median
14%
Std dev
46.3%
10th percentile
69.2%
90th percentile
5.2%
Min
5.2%97.7%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 3.48 for chat models, measured over 92 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#Model or systemScore
1Inkling 2 Small· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim97.7%
2Claude Fable 5.2· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim93%
3Claude Opus 5.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim89.9%
4Claude Opus 5.5Opus 5.5 System Card capability section· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.89.9%
5Claude Sonnet 5.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim81.3%
6Claude Sonnet 5.5adaptive thinking · max effort · default sampling · 5-trial mean · context ≤1M unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.81.3%
7Claude Fable 5.1Opus 5.5 System Card capability section· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.81.2%
8Claude Fable 5.1System Card capability summary · adaptive thinking · max effort · default sampling · 5 trials · context ≤1M unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.81.2%
9Claude Fable 5max· Reported by SpaceXAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.80.4%
10Claude Fable 5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim80%
11Claude Fable 5Fable/Mythos 5 System Card capability summary · adaptive thinking · max effort · 5 trials unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.80%
12Claude Fable 5OpenAI GPT-5.6 launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.80%
13Claude Fable 5Opus 5 System Card capability summary · adaptive thinking · max effort unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.80%
14Claude Fable 5Qwen表:Claude Code · temp 1.0 · top_p 0.95 · 256K · refined tasks· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.80%
15Claude Fable 5System Card capability summary · adaptive thinking · max effort · default sampling · 5 trials · context ≤1M unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.80%
16Claude Opus 5Opus 5 System Card capability summary · adaptive thinking · max effort unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.79.2%
17Claude Opus 5Opus 5.5 System Card capability section· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.79.2%
18Claude Opus 5System Card capability summary · adaptive thinking · max effort · default sampling · 5 trials · context ≤1M unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.79.2%
19Claude Opus 5Atria Dawn Preview official card / report Table 1 · published comparison run · model snapshots and reasoning tiers not specified unless stated · OpenHands v1.23.0 · 4.5-hour limit · 250 interaction steps · network disabled and Git history removed· Reported by Shanghai AI Laboratory· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.74.7%
20Fugu UltraFugu official headline evaluation· Reported by Sakana AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.73.7%
21Claude Opus 4.8· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim69.2%
22Claude Opus 4.8Fable/Mythos 5 System Card capability summary · adaptive thinking · max effort · 5 trials unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.69.2%
23Claude Opus 4.8max· Reported by SpaceXAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.69.2%
24Claude Opus 4.8max · Scale AI · mini-swe-agent for Muse Spark 1.1· Reported by Meta, Meta Superintelligence Labs· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.69.2%
25Claude Opus 4.8OpenAI GPT-5.6 launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.69.2%
26Claude Opus 4.8Opus 4.8 System Card capability summary · adaptive thinking · max effort · 5 trials unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.69.2%
27Claude Opus 4.8Opus 5 System Card capability summary · adaptive thinking · max effort unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.69.2%
28Claude Opus 4.8Qwen表:Claude Code · temp 1.0 · top_p 0.95 · 256K · refined tasks· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.69.2%
29Ling 3.1 Flash· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim68.7%
30Qwen3.8 Max· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
67.7%
31Qwen3.8 MaxQwen表:Claude Code · temp 1.0 · top_p 0.95 · 256K · refined tasks· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.67.7%
32Beam 1.0 501B A23B· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
66.1%
33Hy4 preview· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
65.7%
34Hy4 previewHF Eval Result· Reported by Tencent Hy· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.65.7%
35Qwen3.8 MaxAtria Dawn Preview official card / report Table 1 · published comparison run · model snapshots and reasoning tiers not specified unless stated · OpenHands v1.23.0 · 4.5-hour limit · 250 interaction steps · network disabled and Git history removed· Reported by Shanghai AI Laboratory· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.65.1%
36Claude Haiku 5.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim64.8%
37Claude Haiku 5.5adaptive thinking · max effort · default sampling · 5-trial mean · context ≤1M unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.64.8%
38Grok 4.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim64.7%
39Grok 4.5Gemini 3.6 Flash Model Card · July 2026 · Public· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.64.7%
40Grok 4.5resolve rate· Reported by SpaceXAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.64.7%
41GPT-5.6 Sol· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim64.6%
42GPT-5.6 SolOpenAI GPT-5.6 launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.64.6%
43GPT-5.6 SolOpus 5 System Card capability summary · adaptive thinking · max effort unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.64.6%
44GPT-5.6 SolQwen表:Claude Code · temp 1.0 · top_p 0.95 · 256K · refined tasks· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.64.6%
45GPT-5.6 SolSystem Card capability summary · adaptive thinking · max effort · default sampling · 5 trials · context ≤1M unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.64.6%
46Claude Opus 4.7· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim64.3%
47Claude Opus 4.7adaptive thinking · max effort · avg@5· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.64.3%
48Claude Opus 4.7Gemini 3.5 Flash Model Card · May 2026 · Public · single attempt· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.64.3%
49Claude Opus 4.7max· Reported by SpaceXAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.64.3%
50Claude Opus 4.7Opus 4.8 System Card capability summary · adaptive thinking · max effort · 5 trials unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.64.3%

Showing 1–50 of 187 results.

Not ranked

The source marks these results as descriptive, not capability scores, so we list them by name instead of ranking them.

Model or systemScore
Muse Spark (version unspecified)Meta Model API · highest effort · version unspecified · Scale AI · mini-swe-agent for Muse Spark 1.1· Reported by Meta, Meta Superintelligence Labs· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.55%

Data credits: Hugging Face official benchmark leaderboards; licence varies by benchmark · Data by LLM Stats · BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links. · OpenEvals/leaderboard-data, MIT (licence)