Code migration (Vals AI) leaderboard
- Claude Sonnet 5.5 leads
- 69.8%
- models measured
- 76
- lead over #2
- +1.7 pts
- median
- 31.0%
Leader on Code migration (Vals AI) · read October 21
# Code migration (Vals AI), read October 21
leaders = {
"Claude Sonnet 5.5": 69.8%, "Gemini 4 Argon": 68.2%, "GPT-6 Astra": 67.7%,
}
On October 21, 2026, 76 of our catalog's models had a Code migration (Vals AI) result, and Anthropic Claude Sonnet 5.5 led with 69.8%, ahead of Google Gemini 4 Argon at 68.2%.
Vals AI's code migration board, run with one shared harness; its page describes the tasks. The score as Vals AI publishes it.
Role in the index
A coding board we show beside the Vibe Coding Index, not inside it: the index keeps the six boards it was defined on, so a score does not move every time a board is added. Read the methodology
| # | Model | Result | Cost per test | Duration, as printed | Evidence | Coding index | Output $/Mtok |
|---|---|---|---|---|---|---|---|
| 1 | Claude Sonnet 5.5Anthropic | 69.8% | $75.83 | 3h 20m | 87.3 | $10.00 | |
| 2 | Gemini 4 ArgonGoogle | 68.2% | $57.82 | 2h 12m | 81.0 | $10.00 | |
| 3 | GPT-6 AstraOpenAI | 67.7% | $44.36 | 58m 30s | 85.1 | $50.00 | |
| 4 | Claude Opus 5.5Anthropic | 66.7% | $112.97 | 3h 19m | 88.8 | $20.00 | |
| 5 | GPT-6.1 SolOpenAI | 65.1% | $6.51 | 1h 36m | 82.4 | $10.00 | |
| 6 | Muse WonderMeta | 62.5% | $69.54 | 2h 51m | 71.9 | $20.00 | |
| 7 | Claude Opus 5Anthropic | 57.5% | $60.51 | 2h 51m | 83.0 | $25.00 | |
| 8 | GPT-6 SolOpenAI | 57.2% | $15.68 | 1h 26m | 76.0 | $10.00 | |
| 9 | Claude Fable 5Anthropic | 55.1% | $112.10 | 1h 49m | 79.3 | $50.00 | |
| 10 | Claude Fable 5.1Anthropic | 54.6% | $70.97 | 4h 20m | 84.2 | $50.00 | |
| 11 | GPT-5.6 SolOpenAI | 52.9% | $24.54 | 59m 12s | 73.5 | $20.00 | |
| 12 | GPT-5.6 TerraOpenAI | 47.8% | $8.13 | 48m 34s | 63.9 | $12.00 | |
| 13 | Hy4 previewTencent | 47.4% | $3.41 | 2h 13m | 44.3 | $2.2509 | |
| 14 | Claude Opus 4.8Anthropic | 47.3% | $30.51 | 1h 15m | 65.0 | $25.00 | |
| 15 | DeepSeek V4.1 FlashDeepSeek | 45.6% | $0.94 | 1h 45m | 62.5 | $0.78 | |
| 16 | GPT-5.5OpenAI | 45.2% | $6.44 | 31m 35s | 71.3 | $30.00 | |
| 17 | Grok 4.7xAI | 44.8% | $36.55 | 1h 09m | 66.7 | $6.00 | |
| 18 | Claude Haiku 5.5Anthropic | 44.6% | $6.23 | 1h 23m | 70.7 | $0.50 | |
| 19 | Grok 4.6xAI | 44.6% | $14.58 | 1h 30m | 65.7 | $6.00 | |
| 20 | GPT-5.6 LunaOpenAI | 44.5% | $1.88 | 58m 35s | 58.0 | $1.20 | |
| 21 | Claude Sonnet 5Anthropic | 44.4% | $35.31 | 1h 57m | 57.7 | $10.00 | |
| 22 | GLM 5.3Z.ai | 44.2% | $24.91 | 3h 47m | 70.6 | $4.40 | |
| 23 | DeepSeek V4.1 Flash 1017DeepSeek | 44.2% | $1.86 | 2h 11m | 70.3 | $0.60 | |
| 24 | Claude Opus 4.7Anthropic | 43.9% | $36.39 | 1h 12m | 67.5 | $25.00 | |
| 25 | MiMo-V2.6-ProXiaomi | 43.0% | $0.69 | 2h 49m | 65.4 | $0.87 | |
| 26 | GPT-6 LunaOpenAI | 42.5% | $0.60 | 50m 23s | 57.3 | $0.50 | |
| 27 | DeepSeek V4 Pro 0813DeepSeek | 41.5% | $18.58 | 3h 32m | 56.1 | $1.98 | |
| 28 | MiMo-V2.6-FlashXiaomi | 40.9% | $0.49 | 3h 12m | 54.7 | $0.28 | |
| 29 | Claude Sonnet 4.6Anthropic | 39.9% | $38.51 | 1h 40m | 57.3 | $15.00 | |
| 30 | Step 5 PreviewStepFun | 39.0% | $7.61 | 3h 34m | 61.1 | $2.70 | |
| 31 | DeepSeek V4 Flash 0731DeepSeek | 38.6% | $3.61 | 2h 08m | 51.8 | $0.28 | |
| 32 | GLM 5.2Z.ai | 37.9% | $12.64 | 1h 54m | 61.1 | $4.40 | |
| 33 | Grok 4.5xAI | 36.6% | $3.37 | 23m 29s | 58.6 | $6.00 | |
| 34 | Gemini 3.8 FlashGoogle | 36.5% | $18.49 | 2h 37m | 61.6 | $3.75 | |
| 35 | GPT-5.4OpenAI | 35.0% | $7.76 | 35m 46s | 64.4 | $15.00 | |
| 36 | Gemini 3.7 FlashGoogle | 34.8% | $21.46 | 2h 39m | 58.0 | $3.75 | |
| 37 | Ember-1Fireworks | 32.5% | $51.11 | 4h 09m | 52.6 | $15.00 | |
| 38 | Muse Spark 1.1Meta | 31.1% | $4.56 | 39m 11s | 70.3 | $4.25 | |
| 39 | Gemini 3.6 FlashGoogle | 30.9% | $8.93 | 1h 30m | 62.5 | $3.75 | |
| 40 | Mistral Large 4Mistral AI | 30.6% | $47.89 | 4h 51m | 56.4 | $4.18 | |
| 41 | Muse Spark 1.2Meta | 29.9% | $3.78 | 31m 07s | 55.7 | $4.25 | |
| 42 | Kimi K2.6Moonshot AI | 27.8% | $5.11 | 1h 30m | 52.1 | $4.00 | |
| 43 | Muse Spark 1.3Meta | 27.6% | $4.62 | 1h 08m | 57.4 | $4.25 | |
| 44 | Gemini 3.5 FlashGoogle | 26.8% | $5.73 | 19m 28s | 39.8 | $9.00 | |
| 45 | DeepSeek V4 Pro 0423DeepSeek | 26.2% | $1.07 | 28m 58s | 38.7 | $2.716 | |
| 46 | GLM 5.1Z.ai | 25.8% | $5.70 | 58m 22s | 49.2 | $4.40 | |
| 47 | Kimi K2.7 CodeMoonshot AI | 25.4% | $5.04 | 1h 39m | 53.4 | $4.00 | |
| 48 | Qwen3.8 Max (0902)Qwen | 24.0% | $10.54 | 4h 28m | 62.4 | $6.00 | |
| 49 | MiMo-V2.5-ProXiaomi | 21.6% | $0.24 | 48m 53s | 31.6 | $0.9152 | |
| 50 | Muse Spark OpenMeta | 21.2% | $9.04 | 1h 53m | 47.4 | $4.25 | |
| 51 | GLM 5.3 FlashZ.ai | 20.5% | $3.13 | 3h 26m | 57.7 | $0.50 | |
| 52 | MiniMax M3MiniMax | 19.9% | $7.07 | 1h 14m | 39.2 | $1.20 | |
| 53 | Gemini 3.1 Pro PreviewGoogle | 17.3% | $1.61 | 12m 01s | 32.3 | $12.00 | |
| 54 | Kimi K3Moonshot AI | 16.1% | $13.87 | 4h 18m | 62.3 | $15.00 | |
| 55 | GPT-5.4 NanoOpenAI | 14.5% | $0.42 | 26m 34s | 44.8 | $1.25 | |
| 56 | MiMo-V2.5Xiaomi | 14.2% | $0.10 | 37m 55s | 50.0 | $0.336 | |
| 57 | Qwen3.8 27BQwen | 14.2% | $21.68 | 4h 46m | 74.2 | $2.55 | |
| 58 | T Inkling SmallThinking Machines | 13.7% | $0.60 | 20m 33s | 27.3 | $1.20 | |
| 59 | Qwen3.7 MaxQwen | 13.1% | $1.93 | 25m 32s | 54.1 | $4.425 | |
| 60 | GPT-5.4 MiniOpenAI | 12.9% | $1.95 | 28m 53s | 35.8 | $4.50 | |
| 61 | Qwen3.7 PlusQwen | 12.9% | $0.42 | 35m 47s | 33.2 | $1.28 | |
| 62 | T InklingThinking Machines | 11.8% | $3.01 | 29m 34s | 31.5 | $4.05 | |
| 63 | Qwen3.6 PlusQwen | 11.1% | $4.78 | 29m 50s | 44.1 | $1.95 | |
| 64 | MiniMax M2.7MiniMax | 8.7% | $0.59 | 20m 59s | 36.2 | $1.20 | |
| 65 | Claude Haiku 4.5Anthropic | 7.5% | $0.98 | 9m 54s | 35.6 | $5.00 | |
| 66 | Kimi K2.5Moonshot AI | 7.0% | $0.93 | 1h 04m | 38.6 | $2.85 | |
| 67 | Grok 4.3xAI | 6.8% | $0.42 | 3m 00s | 37.4 | $2.50 | |
| 68 | Gemini 3 Flash PreviewGoogle | 6.4% | $0.34 | 4m 44s | 43.7 | $3.00 | |
| 69 | Gemini 3.5 Flash LiteGoogle | 6.2% | $0.53 | 22m 02s | 49.0 | $2.50 | |
| 70 | Mistral Medium 3.5Mistral AI | 5.1% | $18.62 | 3h 38m | 25.5 | $7.50 | |
| 71 | Nemotron 3 UltraNVIDIA | 4.9% | Not published | 1h 31m | 30.0 | $2.40 | |
| 72 | Gemini 3.1 Flash Lite PreviewGoogle | 4.6% | $0.05 | 1m 33s | 22.7 | $1.50 | |
| 73 | Mercury 2.5Inception | 4.5% | $0.63 | 16m 14s | 1.6 | $0.15 | |
| 74 | Nemotron 3.5 LightningNVIDIA | 3.0% | $0.27 | 43m 25s | 22.6 | $0.147 | |
| 75 | Grok 4.20xAI | 0.3% | $0.29 | 6m 14s | 33.3 | $2.50 | |
| 76 | Ling 3.0 FlashinclusionAI | 0.0% | $0.06 | 4h 59m | 29.6 | $0.1215 |
Only models in our catalog are listed, and each appears once with its best published run. 136 rows were read from the source on October 21, 2026, and rows that match no catalog model are left out. Cost per test is what Vals AI reports one test cost on the run behind the result, in US dollars. Every other column is a figure the board publishes for the same run.