BlogLLM Benchmark Scores as Published, Side by Side
SWE-bench Verified: Results as Published
Every result the catalog holds for SWE-bench Verified: 88 models from 15 developers, read from 40 pages. Each number is as printed by the company named beside it; nothing is ranked or rescored. Models are listed by developer, newest first.
Area Agentic CodingMaintainer OpenAI (built on SWE-bench by Princeton University)Introduced 2024Data version 2026103001, as of 2026-10-30
SWE-bench Verified is a human-validated subset of 500 tasks from SWE-bench, the benchmark where a model receives a real GitHub issue from a Python repository and must produce a patch that makes the hidden tests pass. Professional software developers screened the original tasks and kept those with a clear problem statement and fair tests, so a failure reflects the model rather than a broken task.
SWE-bench Verified
| Model | Value | Printed by | Setup as printed |
|---|---|---|---|
| GPT-5.5OpenAI, released 2026-04-23 | 82.9% | MiniMax, MiniMax M3checked 2026-10-15 | Harness: Claude Code |
| GPT-5.2OpenAI, released 2025-12-11 | 80% | OpenAI, Introducing GPT-5.2checked 2026-10-15; printed on 6 more pages | Effort: xhigh |
| GPT-5.1-Codex-MaxOpenAI, released 2025-11-19 | 77.9% | Anthropic, Claude Opus 4.5checked 2026-10-15; printed on 1 more page | - |
| GPT-5.1OpenAI, released 2025-11-12 | 76.3% | OpenAI, Introducing GPT-5.1 for developerschecked 2026-10-15; printed on 3 more pages | Effort: high; Harness: harness with JSON-based apply_patch tool; Dataset revision: all 500 problems |
| GPT-5-CodexOpenAI, released 2025-09-15 | 74.5% | Anthropic, Claude Haiku 4.5checked 2026-10-15; printed on 1 more page | - |
| GPT-5OpenAI, released 2025-08-07 | 52.8% | OpenAI, Introducing GPT-5checked 2026-10-15; printed on 6 more pages | Effort: none |
| 74.9% | OpenAI, Introducing GPT-5checked 2026-10-15 | Effort: high | |
| GPT-5 miniOpenAI, released 2025-08-07 | 72% | Alibaba, Qwen3.5-122B-A10B model cardchecked 2026-10-15 | - |
| gpt-oss-120bOpenAI, released 2025-08-05 | 62% | Alibaba, Qwen3.5-122B-A10B model cardchecked 2026-10-15; printed on 1 more page | - |
| gpt-oss-20bOpenAI, released 2025-08-05 | 52.44% | NVIDIA, NVIDIA Nemotron 3.5 Lightning 30B-A3B model cardchecked 2026-10-19 | - |
| codex-1OpenAI, released 2025-05-16 | 72.1% | Anthropic, Claude Opus 4 and Claude Sonnet 4checked 2026-10-15 | - |
| o3OpenAI, released 2025-04-16 | 69.1% | OpenAI, Introducing GPT-5checked 2026-10-15; printed on 2 more pages | Effort: high |
| GPT-4.1OpenAI, released 2025-04-14 | 54.6% | Anthropic, Claude Opus 4 and Claude Sonnet 4checked 2026-10-15; printed on 1 more page | - |
| o3-miniOpenAI, released 2025-01-31 | 49.3% | Anthropic, Claude 3.7 Sonnetchecked 2026-10-15 | Effort: high |
| o1OpenAI, released 2024-12-17 | 48.9% | Anthropic, Claude 3.7 Sonnetchecked 2026-10-15; printed on 1 more page | - |
| o1-miniOpenAI, released 2024-09-12 | 41.6% | DeepSeek, DeepSeek-R1 model cardchecked 2026-10-15 | - |
| GPT-4oOpenAI, released 2024-05-13 | 30.8% | OpenAI, Introducing GPT-5checked 2026-10-15; printed on 1 more page | - |
| Claude Fable 5Anthropic, released 2026-06-09 | 95% | Thinking Machines, Thinking Machines Inkling model cardchecked 2026-10-15 | Effort: max |
| Claude Opus 4.7Anthropic, released 2026-04-16 | 87.6% | Anthropic, Claude Opus 4.7checked 2026-10-15; printed on 1 more page | - |
| Claude Mythos PreviewAnthropic, released 2026-04-07 | 93.9% | Anthropic, Claude Opus 4.7checked 2026-10-15 | - |
| Claude Sonnet 4.6Anthropic, released 2026-02-17 | 79.6% | Anthropic, Claude Sonnet 4.6checked 2026-10-15; printed on 4 more pages | - |
| Claude Opus 4.6Anthropic, released 2026-02-05 | 80.8% | Anthropic, Claude Opus 4.6checked 2026-10-15; printed on 7 more pages | - |
| Claude Opus 4.5Anthropic, released 2025-11-24 | 80.9% | Anthropic, Claude Opus 4.5checked 2026-10-15; printed on 6 more pages | Effort: high |
| Claude Haiku 4.5Anthropic, released 2025-10-15 | 73.3% | Anthropic, Claude Haiku 4.5checked 2026-10-15 | Effort: thinking budget 128K |
| Claude Sonnet 4.5Anthropic, released 2025-09-29 | 78.2% | Anthropic, Claude Sonnet 4.5checked 2026-10-15; printed on 9 more pages | Effort: thinking budget 200K; Context: 1M |
| 82% | Anthropic, Claude Sonnet 4.5checked 2026-10-15 | Effort: thinking budget 200K; Attempt budget: parallel test-time compute: multiple samples, the best selected by an internal scoring model | |
| 77.2% | Anthropic, Claude Sonnet 4.5checked 2026-10-15 | Effort: thinking budget 200K | |
| Claude Opus 4.1Anthropic, released 2025-08-05 | 74.5% | Anthropic, Claude Opus 4.1checked 2026-10-15; printed on 2 more pages | Effort: none |
| Claude Opus 4Anthropic, released 2025-05-22 | 79.4% | Anthropic, Claude Opus 4 and Claude Sonnet 4checked 2026-10-15; printed on 2 more pages | Effort: none; Attempt budget: parallel test-time compute: multiple samples, the best selected by an internal scoring model |
| 72.5% | Anthropic, Claude Opus 4 and Claude Sonnet 4checked 2026-10-15 | Effort: none | |
| Claude Sonnet 4Anthropic, released 2025-05-22 | 80.2% | Anthropic, Claude Sonnet 4.5checked 2026-10-15; printed on 4 more pages | Effort: thinking budget 200K; Attempt budget: parallel test-time compute: multiple samples, the best selected by an internal scoring model |
| 72.7% | Anthropic, Claude Sonnet 4.5checked 2026-10-15 | Effort: thinking budget 200K | |
| Claude 3.7 SonnetAnthropic, released 2025-02-24 | 70.3% | Anthropic, Claude 3.7 Sonnetchecked 2026-10-15; printed on 1 more page | Effort: none; Harness: custom scaffold with internal scoring on a reduced subset of problems |
| 62.3% | Anthropic, Claude 3.7 Sonnetchecked 2026-10-15 | Effort: none | |
| Claude 3.5 SonnetAnthropic, released 2024-10-22 | 49% | Anthropic, Claude 3.7 Sonnetchecked 2026-10-15; printed on 1 more page | - |
| Gemma 4 26B A4BGoogle, released 2026-04-02 | 17.4% | Alibaba, Qwen3.6-35B-A3B model cardchecked 2026-10-15; printed on 1 more page | Harness: internal scaffold; Context: 200000 |
| Gemma 4 31BGoogle, released 2026-04-02 | 52% | Alibaba, Qwen3.6-27B model cardchecked 2026-10-15; printed on 1 more page | Harness: internal scaffold; Context: 200000 |
| Gemini 3.1 Pro PreviewGoogle, released 2026-02-19 | 80.6% | Google, Gemini 3.1 Pro model cardchecked 2026-10-15; printed on 5 more pages | Effort: high |
| Gemini 3 FlashGoogle, released 2025-12-17 | 78% | Anthropic, Claude Sonnet 4.6checked 2026-10-15 | - |
| Gemini 3 ProGoogle, released 2025-11-18 | 76.2% | Google, Gemini 3.1 Pro model cardchecked 2026-10-15; printed on 8 more pages | Effort: high |
| Gemini 2.5 ProGoogle, released 2025-03-25 | 63.2% | Anthropic, Claude Opus 4 and Claude Sonnet 4checked 2026-10-15; printed on 3 more pages | - |
| Grok Code Fast 1xAI, released 2025-08-28 | 70.8% | Mistral AI, Devstral 2 and Mistral Vibe CLIchecked 2026-10-15 | - |
| Code World Model (CWM)Meta, released 2025-09-24 | 53.9% | Mistral AI, Devstral 2 and Mistral Vibe CLIchecked 2026-10-15 | - |
| DeepSeek-V4-Flash (Preview)DeepSeek, released 2026-04-24 | 79% | DeepSeek, DeepSeek-V4-Pro (preview) model cardchecked 2026-10-15; printed on 1 more page | Effort: max |
| 73.7% | DeepSeek, DeepSeek-V4-Pro (preview) model cardchecked 2026-10-15 | Effort: non-think | |
| 78.6% | DeepSeek, DeepSeek-V4-Pro (preview) model cardchecked 2026-10-15 | Effort: high | |
| DeepSeek-V4-Pro (Preview)DeepSeek, released 2026-04-24 | 79.4% | DeepSeek, DeepSeek-V4-Pro (preview) model cardchecked 2026-10-15; printed on 4 more pages | Effort: high |
| 73.6% | DeepSeek, DeepSeek-V4-Pro (preview) model cardchecked 2026-10-15 | Effort: non-think | |
| 80.6% | DeepSeek, DeepSeek-V4-Pro (preview) model cardchecked 2026-10-15 | Effort: max | |
| DeepSeek-V3.2DeepSeek, released 2025-12-01 | 73.1% | DeepSeek, DeepSeek-V3.2 model cardchecked 2026-10-15; printed on 5 more pages | Effort: thinking |
| DeepSeek-V3.2-ExpDeepSeek, released 2025-09-29 | 67.8% | DeepSeek, DeepSeek-V3.2-Exp model cardchecked 2026-10-15; printed on 1 more page | Tools: enabled; exact tools in source |
| DeepSeek-V3.1-TerminusDeepSeek, released 2025-09-22 | 68.4% | DeepSeek, DeepSeek-V3.1-Terminus model cardchecked 2026-10-15; printed on 1 more page | Tools: enabled; exact tools in source |
| DeepSeek-V3.1DeepSeek, released 2025-08-21 | 66% | DeepSeek, DeepSeek-V3.1-Terminus model cardchecked 2026-10-15 | Tools: enabled; exact tools in source |
| DeepSeek-R1-0528DeepSeek, released 2025-05-28 | 57.6% | DeepSeek, DeepSeek-R1-0528 model cardchecked 2026-10-15 | Harness: Agentless |
| DeepSeek-V3-0324DeepSeek, released 2025-03-24 | 38.8% | Moonshot AI, Kimi K2 release pagechecked 2026-10-15 | Harness: agentic, single attempt |
| DeepSeek-R1DeepSeek, released 2025-01-20 | 49.2% | DeepSeek, DeepSeek-R1 model cardchecked 2026-10-15; printed on 2 more pages | - |
| DeepSeek-V3DeepSeek, released 2024-12-26 | 42% | DeepSeek, DeepSeek-R1 model cardchecked 2026-10-15 | - |
| DeepSeek-V4-Flash-0731DeepSeek | 75.4% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
| Qwen3.6-27BAlibaba, released 2026-04-22 | 77.2% | Alibaba, Qwen3.6-27B model cardchecked 2026-10-15 | Harness: internal scaffold; Context: 200000 |
| Qwen3.6-35B-A3BAlibaba, released 2026-04-16 | 73.4% | Alibaba, Qwen3.6-35B-A3B model cardchecked 2026-10-15; printed on 2 more pages | Harness: internal scaffold; Context: 200000 |
| Qwen3.5-122B-A10BAlibaba, released 2026-02-24 | 72% | Alibaba, Qwen3.5-122B-A10B model cardchecked 2026-10-15; printed on 1 more page | - |
| Qwen3.5-27BAlibaba, released 2026-02-24 | 72.4% | Alibaba, Qwen3.5-122B-A10B model cardchecked 2026-10-15; printed on 2 more pages | - |
| Qwen3.5-35B-A3BAlibaba, released 2026-02-24 | 69.2% | Alibaba, Qwen3.5-122B-A10B model cardchecked 2026-10-15; printed on 1 more page | - |
| Qwen3.5-397B-A17BAlibaba, released 2026-02-16 | 76.4% | Alibaba, Qwen3.5-397B-A17B model cardchecked 2026-10-15; printed on 3 more pages | - |
| Qwen3-Coder-PlusAlibaba, released 2025-07-23 | 69.6% | Mistral AI, Devstral 2 and Mistral Vibe CLIchecked 2026-10-15 | - |
| Qwen3-235B-A22BAlibaba, released 2025-04-29 | 34.4% | Moonshot AI, Kimi K2 release pagechecked 2026-10-15 | Effort: non-thinking; Harness: agentic, single attempt |
| Qwen3-Max-ThinkingAlibaba | 75.3% | Alibaba, Qwen3.5-397B-A17B model cardchecked 2026-10-15 | - |
| Qwen3.8-Flash-NextAlibaba | 78.7% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
| Kimi K2.6Moonshot AI, released 2026-04-20 | 80.2% | Moonshot AI, Kimi K2.6 model cardchecked 2026-10-15; printed on 5 more pages | Harness: in-house SWE-agent variant |
| Kimi K2.5Moonshot AI, released 2026-01-27 | 76.8% | Moonshot AI, Kimi K2.5 model cardchecked 2026-10-15; printed on 5 more pages | Harness: in-house SWE-agent variant |
| Kimi K2 ThinkingMoonshot AI, released 2025-11-06 | 71.3% | Moonshot AI, Kimi K2 Thinking release pagechecked 2026-10-15; printed on 2 more pages | Effort: thinking; Harness: in-house harness derived from SWE-agent; Tools: enabled; bash and edit tools |
| Kimi K2 Instruct 0905Moonshot AI, released 2025-09-05 | 69.2% | Moonshot AI, Kimi K2 Thinking release pagechecked 2026-10-15 | Harness: in-house harness derived from SWE-agent; Tools: enabled; bash and edit tools |
| Kimi K2 InstructMoonshot AI, released 2025-07-11 | 65.8% | Moonshot AI, Kimi K2 release pagechecked 2026-10-15 | Harness: agentic, single attempt |
| GLM-5.3-FlashZ.ai, released 2026-08-26 | 79.1% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
| GLM-5.1Z.ai, released 2026-04-07 | 80.2% | Mistral AI, Mistral Medium 3.5 model cardchecked 2026-10-15; printed on 1 more page | Effort: max |
| GLM-5Z.ai, released 2026-02-12 | 77.8% | Z.ai, GLM-5 model cardchecked 2026-10-15 | Harness: OpenHands; Context: 200000 |
| GLM-4.7Z.ai, released 2025-12-22 | 73.8% | Z.ai, GLM-4.7 model cardchecked 2026-10-15; printed on 1 more page | - |
| GLM-4.6Z.ai, released 2025-09-30 | 68% | Z.ai, GLM-4.7 model cardchecked 2026-10-15; printed on 1 more page | - |
| Mistral Medium 3.5Mistral AI, released 2026-05-22 | 77.6% | Mistral AI, Mistral Medium 3.5 model cardchecked 2026-10-15 | Effort: max |
| Devstral 2Mistral AI, released 2025-12-09 | 72.2% | Mistral AI, Devstral 2 and Mistral Vibe CLIchecked 2026-10-15; printed on 1 more page | - |
| Devstral Small 2Mistral AI, released 2025-12-09 | 68% | Mistral AI, Devstral 2 and Mistral Vibe CLIchecked 2026-10-15; printed on 1 more page | - |
| MiniMax M3MiniMax, released 2026-06-01 | 80.5% | MiniMax, MiniMax M3checked 2026-10-15 | Harness: Claude Code |
| MiniMax M2.7MiniMax, released 2026-03-18 | 79.9% | MiniMax, MiniMax M3checked 2026-10-15; printed on 1 more page | Harness: Claude Code |
| MiniMax M2.5MiniMax, released 2026-02-12 | 79.7% | MiniMax, MiniMax-M2.5 model cardchecked 2026-10-15 | Harness: Droid scaffold, default prompt |
| 76.1% | MiniMax, MiniMax-M2.5 model cardchecked 2026-10-15 | Harness: OpenCode scaffold, default prompt | |
| MiniMax M2MiniMax, released 2025-10-27 | 69.4% | Mistral AI, Devstral 2 and Mistral Vibe CLIchecked 2026-10-15 | - |
| Nemotron 3.5 Super VL 120B-A12BNVIDIA, released 2026-10-15 | 71.1% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
| Nemotron 3.5 Lightning 30B-A3BNVIDIA, released 2026-08-11 | 51.56% | NVIDIA, NVIDIA Nemotron 3.5 Lightning 30B-A3B model cardchecked 2026-10-19 | - |
| Nemotron 3 UltraNVIDIA, released 2026-06-04 | 73.9% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19; printed on 2 more pages | - |
| Nemotron 3 SuperNVIDIA, released 2026-03-11 | 63.08% | NVIDIA, NVIDIA Nemotron 3.5 Lightning 30B-A3B model cardchecked 2026-10-19 | - |
| Nemotron 3 Nano 30B-A3BNVIDIA, released 2025-12-15 | 34.08% | NVIDIA, NVIDIA Nemotron 3.5 Lightning 30B-A3B model cardchecked 2026-10-19 | - |
| MAI-Thinking-1Microsoft, released 2026-08-12 | 73.5% | Microsoft, Microsoft AI: Introducing MAI-Thinking-1checked 2026-10-19 | - |
| InklingThinking Machines, released 2026-07-14 | 77.6% | Thinking Machines, Thinking Machines Inkling model cardchecked 2026-10-15 | Effort: 0.99 |
| Inkling-SmallThinking Machines | 75.27% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
SWE-bench Verified (agentless, single patch without test)
| Model | Value | Printed by | Setup as printed |
|---|---|---|---|
| GPT-4.1OpenAI, released 2025-04-14 | 40.8% | Moonshot AI, Kimi K2 release pagechecked 2026-10-15 | Harness: agentless |
| Claude Opus 4Anthropic, released 2025-05-22 | 53% | Moonshot AI, Kimi K2 release pagechecked 2026-10-15 | Effort: no extended thinking; Harness: agentless |
| Claude Sonnet 4Anthropic, released 2025-05-22 | 50.2% | Moonshot AI, Kimi K2 release pagechecked 2026-10-15 | Effort: no extended thinking; Harness: agentless |
| Gemini 2.5 FlashGoogle, released 2025-05-20 | 32.6% | Moonshot AI, Kimi K2 release pagechecked 2026-10-15 | Harness: agentless |
| DeepSeek-V3-0324DeepSeek, released 2025-03-24 | 36.6% | Moonshot AI, Kimi K2 release pagechecked 2026-10-15 | Harness: agentless |
| Qwen3-235B-A22BAlibaba, released 2025-04-29 | 39.4% | Moonshot AI, Kimi K2 release pagechecked 2026-10-15 | Effort: non-thinking; Harness: agentless |
| Kimi K2 InstructMoonshot AI, released 2025-07-11 | 51.8% | Moonshot AI, Kimi K2 release pagechecked 2026-10-15 | Harness: agentless |
SWE-bench (agentless, split not stated)
| Model | Value | Printed by | Setup as printed |
|---|---|---|---|
| Kimi K2 BaseMoonshot AI, released 2025-07-11 | 28.2% | Xiaomi, Xiaomi MiMo-V2.5-Pro model cardchecked 2026-10-15 | Harness: 3-shot |
| MiMo-V2.5 BaseXiaomi, released 2026-04-27 | 30.8% | Xiaomi, Xiaomi MiMo-V2.5-Pro model cardchecked 2026-10-15 | Harness: 3-shot |
| MiMo-V2.5-Pro BaseXiaomi, released 2026-04-27 | 35.7% | Xiaomi, Xiaomi MiMo-V2.5-Pro model cardchecked 2026-10-15 | Harness: 3-shot |
SWE-bench (split not stated)
| Model | Value | Printed by | Setup as printed |
|---|---|---|---|
| gpt-oss-20bOpenAI, released 2025-08-05 | 34% | NVIDIA, NVIDIA Nemotron 3 Nano 30B-A3B model cardchecked 2026-10-15 | Harness: OpenHands |
| Qwen3-30B-A3B-Thinking-2507Alibaba, released 2025-07-30 | 22% | NVIDIA, NVIDIA Nemotron 3 Nano 30B-A3B model cardchecked 2026-10-15 | Effort: thinking; Harness: OpenHands |
| Nemotron 3 Nano 30B-A3BNVIDIA, released 2025-12-15 | 38.8% | NVIDIA, NVIDIA Nemotron 3 Nano 30B-A3B model cardchecked 2026-10-15 | Harness: OpenHands |
SWE-bench Verified (multiple attempts)
| Model | Value | Printed by | Setup as printed |
|---|---|---|---|
| Claude Opus 4Anthropic, released 2025-05-22 | 79.4% | Moonshot AI, Kimi K2 release pagechecked 2026-10-15 | Effort: no extended thinking; Harness: agentic, multiple attempts |
| Claude Sonnet 4Anthropic, released 2025-05-22 | 80.2% | Moonshot AI, Kimi K2 release pagechecked 2026-10-15 | Effort: no extended thinking; Harness: agentic, multiple attempts |
| Kimi K2 InstructMoonshot AI, released 2025-07-11 | 71.6% | Moonshot AI, Kimi K2 release pagechecked 2026-10-15 | Harness: agentic, multiple attempts |
Every number here is read from the same catalog file as the benchmark table, data version 2026103001, and is the number the table shows for that model and test. Where one page printed a model under several setups, each value is listed. The catalog is public in armanobosyan/sugra-llm-benchmarks; a wrong number can be reported from the table, with Report an error under the value, or as an issue in that repository.