BlogLLM Benchmark Scores as Published, Side by Side
Kimi Code Bench: Results as Published
Every result the catalog holds for Kimi Code Bench: 14 models from 5 developers, read from 3 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 Moonshot AIIntroduced 2026Data version 2026103001, as of 2026-10-30
Kimi Code Bench is an in-house agentic coding benchmark of Moonshot AI, run inside a coding agent harness. Version 2.0 has 80 tasks, and the set includes cybersecurity and safety-related tasks, so the publisher also discloses how many tasks a model refused or handled with a fallback. The card prints one number per model and does not name the metric.
Kimi Code Bench 2.0
| Model | Value | Printed by | Setup as printed |
|---|---|---|---|
| GPT-5.6 SolOpenAI, released 2026-07-09 | 64.8% | Moonshot AI, Kimi K3 model cardchecked 2026-10-15 | Effort: max; Harness: Codex |
| GPT-5.5OpenAI, released 2026-04-23 | 69% | Moonshot AI, Kimi K2.7 Code model cardchecked 2026-10-15; printed on 1 more page | Effort: xhigh; Context: 262144 |
| Claude Fable 5Anthropic, released 2026-06-09 | 76.9% | Moonshot AI, Kimi K3 model cardchecked 2026-10-15 | Effort: max; Harness: Claude Code; Fallback: w/ fallback |
| Claude Opus 4.8Anthropic, released 2026-05-28 | 67.4% | Moonshot AI, Kimi K2.7 Code model cardchecked 2026-10-15; printed on 1 more page | Effort: xhigh; Context: 262144 |
| DeepSeek-V4.1-Flash-1017DeepSeek, released 2026-10-17 | 76.1% | Moonshot AI, Introducing Kimi K3.1 Prochecked 2026-10-24 | Effort: max |
| Kimi K3.1 ProMoonshot AI, released 2026-10-23 | 77.1% | Moonshot AI, Introducing Kimi K3.1 Prochecked 2026-10-24 | Effort: max; Harness: Claude Code |
| Kimi K3Moonshot AI, released 2026-07-16 | 72.9% | Moonshot AI, Kimi K3 model cardchecked 2026-10-15 | Effort: max; Harness: Kimi Code |
| 73.7% | Moonshot AI, Kimi K3 model cardchecked 2026-10-15 | Effort: max; Harness: Claude Code | |
| Kimi K2.7 CodeMoonshot AI, released 2026-06-11 | 62% | Moonshot AI, Kimi K2.7 Code model cardchecked 2026-10-15 | Context: 262144 |
| Kimi K2.6Moonshot AI, released 2026-04-20 | 50.9% | Moonshot AI, Kimi K2.7 Code model cardchecked 2026-10-15 | Context: 262144 |
| GLM-5.3Z.ai, released 2026-08-18 | 61.7% | Moonshot AI, Introducing Kimi K3.1 Prochecked 2026-10-24 | Effort: max |
| GLM-5.2Z.ai, released 2026-06-16 | 64.2% | Moonshot AI, Kimi K3 model cardchecked 2026-10-15 | Effort: max; Harness: Claude Code |
Kimi Code Bench 3.0 Long-horizon
| Model | Value | Printed by | Setup as printed |
|---|---|---|---|
| GPT-6 AstraOpenAI, released 2026-09-03 | 66.5% | Moonshot AI, Introducing Kimi K3.1 Prochecked 2026-10-24 | Effort: max; Fallback: with fallback |
| Claude Opus 5.5Anthropic, released 2026-09-22 | 58.5% | Moonshot AI, Introducing Kimi K3.1 Prochecked 2026-10-24 | Effort: max; Fallback: with fallback |
| Claude Fable 5.1Anthropic, released 2026-09-01 | 58.8% | Moonshot AI, Introducing Kimi K3.1 Prochecked 2026-10-24 | Effort: max; Fallback: with fallback |
| DeepSeek-V4.1-Flash-1017DeepSeek, released 2026-10-17 | 51.5% | Moonshot AI, Introducing Kimi K3.1 Prochecked 2026-10-24 | Effort: max |
| Kimi K3.1 ProMoonshot AI, released 2026-10-23 | 61.2% | Moonshot AI, Introducing Kimi K3.1 Prochecked 2026-10-24 | Effort: max; Harness: Claude Code |
| Kimi K3Moonshot AI, released 2026-07-16 | 55.95% | Moonshot AI, Introducing Kimi K3.1 Prochecked 2026-10-24 | - |
| GLM-5.3Z.ai, released 2026-08-18 | 45.5% | Moonshot AI, Introducing Kimi K3.1 Prochecked 2026-10-24 | Effort: max |
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.