DeepSWE v1.1
An AI capability benchmark for software engineering tasks, evaluating a model's code comprehension, refactoring, and defect-fixing performance in real production repositories.
Core Facts
DeepSeek-V4.1-Flash-1017 scores 77.0 on DeepSWE v1.1, surpassing Claude Opus 5's 74.0.
V4.1-Flash-1017 scores 77.0 on the DeepSWE benchmark.
The OpenRouter model pages show that across all traffic, DeepSeek V4.1 Flash accounts for about 36% of tokens and GPT-6.1 Sol about 18%.
Timeline (last 90 days)
Official benchmarks show DeepSWE v1.1 improving from 74.2 to 77.0 (Mini-SWE-V2 environment), and Terminal-Bench 4.0 improving from 31.2 to 42.3 (Claude Code environment).
According to official self-reported data, compared with V4.1-Flash, V4.1-Flash-1017 improved on Terminal-Bench 4.0 from 31.2 to 42.3 (+11.1), on HLE from 36.8 to 42.6 (+5.8), on DeepSWE v1.1 from 74.2 to 77.0 (+2.8), and on Automation-Bench from 54.8 to 60.5 (+5.7).
DeepSeek-V4.1-Flash-1017 scores 77.0 on DeepSWE v1.1, surpassing Claude Opus 5's 74.0.
The OpenRouter model pages show that across all traffic, DeepSeek V4.1 Flash accounts for about 36% of tokens and GPT-6.1 Sol about 18%.
V4.1-Flash-1017 scores 77.0 on the DeepSWE benchmark.
Gemini 4 Argon achieved a SOTA score of 77.9% on the DeepSWE v1.1 benchmark.
Argon leads on DeepSWE v1.1 with 77.9% versus Opus 5.5's 74.2%; on Terminal-Bench 4.0, Opus 5.5 leads with 66.4% versus 57.4%.
On the DeepSWE v1.1 (coding) benchmark, Gemini 4 Argon scores 77.9%, Claude Opus 5.5 scores 74.2%, and GPT-6 Astra scores 74.1%.
GLM-5.3-Flash scores 63.4 on DeepSWE v1.1, a 17.2-point improvement over GLM-5.2's 46.2, and 55.3 on HLE.
In the Deep Study software engineering test, Argon scored 77.9%, higher than GPT-6 Astra and Claude Opus 5.5 (both approximately 74%).
2 more timeline events
All Facts (8)
DeepSeek-V4.1-Flash-1017 scores 77.0 on DeepSWE v1.1, surpassing Claude Opus 5's 74.0.
70%VerifiedV4.1-Flash-1017 scores 77.0 on the DeepSWE benchmark.
65%Partially VerifiedThe OpenRouter model pages show that across all traffic, DeepSeek V4.1 Flash accounts for about 36% of tokens and GPT-6.1 Sol about 18%.
65%UnverifiedOfficial benchmarks show DeepSWE v1.1 improving from 74.2 to 77.0 (Mini-SWE-V2 environment), and Terminal-Bench 4.0 improving from 31.2 to 42.3 (Claude Code environment).
50%UnverifiedAccording to official self-reported data, compared with V4.1-Flash, V4.1-Flash-1017 improved on Terminal-Bench 4.0 from 31.2 to 42.3 (+11.1), on HLE from 36.8 to 42.6 (+5.8), on DeepSWE v1.1 from 74.2 to 77.0 (+2.8), and on Automation-Bench from 54.8 to 60.5 (+5.7).
50%UnverifiedGemini 4 Argon achieved a SOTA score of 77.9% on the DeepSWE v1.1 benchmark.
50%UnverifiedArgon set a new record of 77.9% on DeepSWE v1.1, which measures real long-horizon software engineering tasks.
50%UnverifiedOn the DeepSWE v1.1 benchmark, 3.8 Flash surpasses most frontier models with larger parameter counts.
50%