The Inference Report
July 26, 2026
The Value FrontierDaily rankings from SWE-rebench, a benchmark designed to fairly compare LLM capabilities on real-world software engineering tasks. Unlike other evaluations, it uses a standardized scaffolding for all models, continuously updates its dataset to prevent contamination, and runs each model five times to account for stochastic variance.
| # | Model | Score |
|---|---|---|
| 1 | OpenAIgpt-5.5-2026-04-23-xhighModel | 62.7%± 0.91% |
| 2 | JunieJunieAgent | 61.6%± 0.64% |
| 3 | OpenAICodexAgent | 60.4%± 1.37% |
| 4 | AnthropicClaude CodeAgent | 59.6%± 1.98% |
| 5 | OpenAIgpt-5.5-2026-04-23-mediumModel | 58.9%± 0.78% |
| 6 | AnthropicClaude Opus 4.8-xhighModel | 56.5%± 1.20% |
| 7 | OpenAIgpt-5.4-2026-03-05-mediumModel | 54.9%± 1.02% |
| 8 | AnthropicClaude Opus 4.7-highModel | 53.1%± 1.45% |
| 9 | CursorCursorAgent | 53.0%± 0.53% |
| 10 | AnthropicClaude Sonnet 4.6Model | 51.3%± 0.55% |
Artificial Analysis composite index across coding, math, and reasoning benchmarks.
| # | Model | Score | tok/s | $/1M |
|---|---|---|---|---|
| 1 | Claude Opus 5 | 60.7 | 44 | $10.00 |
| 2 | Claude Fable 5 | 59.9 | 58 | $20.00 |
| 3 | GPT-5.6 Sol | 58.9 | 74 | $11.25 |
| 4 | Kimi K3 | 57.1 | 33 | $6.00 |
| 5 | Claude Opus 4.8 | 55.7 | 63 | $10.00 |
| 6 | GPT-5.6 Terra | 55 | 128 | $5.63 |
| 7 | GPT-5.5 | 54.8 | 0 | $11.25 |
| 8 | Grok 4.5 | 53.8 | 56 | $3.00 |
| 9 | Claude Opus 4.7 | 53.5 | 0 | $10.00 |
| 10 | Claude Sonnet 5 | 53.4 | 83 | $4.00 |
Output tokens per second — higher is faster. Minimum intelligence score of 40.
| # | Model | tok/s |
|---|---|---|
| 1 | Gemini 3.5 Flash | 250 |
| 2 | Gemini 3.6 Flash | 219 |
| 3 | Qwen3.7 Max | 200 |
| 4 | GPT-5.6 Luna | 171 |
| 5 | GLM-5.2 | 157 |
| 6 | Gemini 3.1 Pro Preview | 132 |
| 7 | Nex-N2-Pro | 129 |
| 8 | GPT-5.6 Terra | 128 |
| 9 | GPT-5.3 Codex | 126 |
| 10 | Muse Spark 1.1 | 124 |
Blended cost per 1M tokens (3:1 input/output) — lower is cheaper. Minimum intelligence score of 40.
| # | Model | $/1M |
|---|---|---|
| 1 | DeepSeek V4 Flash | $0.175 |
| 2 | Hy3 | $0.25 |
| 3 | MiniMax-M3 | $0.525 |
| 4 | DeepSeek V4 Pro | $0.544 |
| 5 | MiMo-V2.5-Pro | $0.544 |
| 6 | Nex-N2-Pro | $1.00 |
| 7 | GPT-5.4 mini | $1.69 |
| 8 | Kimi K2.6 | $1.71 |
| 9 | Kimi K2.7 Code | $1.71 |
| 10 | Muse Spark 1.1 | $2.00 |