Model data
Find the right model for the job
Choose a benchmark below. Higher dots score better; dots farther left cost less on the selected cost basis. Each dot is a model at one thinking level.
The full comparison. Choose a provider above to explore its models.
General intelligence · Reported cost / task
57.6 / 100
Highest tested score
Opus 5.5 · max · $5.98 · reported cost / task
$1.82
Cheapest within 5 points of the top
Opus 5.5 · high · 53.6 / 100 · a budget shortlist, not a proven tie
15 of 139
Worth comparing
No selected option scores higher for the same or lower cost, or costs less for the same score. Small gaps may be noise.
Intelligence Index methodology and results ↗ · 139 configurations shown · 0 source configurations excluded without usable scores or costs. Snapshot captured Sep 29, 2026. API prices checked Sep 28, 2026. Each benchmark uses its own workload and cost basis; compare costs only within a view.
Artificial Analysis Intelligence Index v4.3, captured Sep 29, 2026. Every reasoning level the source publishes is its own dot, including non-reasoning modes. Cost is the source’s published cost to run the index per task at its listed prices; it has no token breakdown, so Lens does not reprice it. Scores the source marks as estimated are left out. The index measures broad capability, not coding alone. Use the coding benchmarks to narrow the choice.
Your session cost vs DeepSWE score
Different workloads, shown for context rather than a controlled comparison: your average cost per session on the horizontal axis, against the pass rate DeepSWE measured for that model and reasoning effort on the vertical axis.
Import your sessions to price your own work
Your agents run longer tasks than a benchmark, or shorter ones, and lean on the cache differently. Import your logs and Lens will plot what an average session costs you on each model.
Advanced: combined Lens coding profile
Lens coding profile · v2
Three benchmark sources, five quality checks. Rank uses 75% quality and 25% estimated DeepSWE cost. The frontier compares quality and cost directly, without the ranking weights.
69 scored configurations across 27 models · 60 provisional estimates. 1 model identities lack pricing.
Each available reasoning level receives its own score and DeepSWE token-cost estimate. Estimated components use a same-model anchor scaled by the DeepSWE effort ratio, or a DeepSWE-only proxy when no anchor exists. This transfer assumption is unvalidated; estimated scores can change the frontier.
Missing prices: qwen3.8-max
Compare tested model/agent configurations, not isolated model ability. Matching model and effort does not make different harnesses equivalent. This profile measures coding-related performance, not universal quality, speed, or your own success rate.
Hollow dots include estimated components; filled dots have all five measured components. Lines connect each model’s available reasoning levels. Quality is an explicitly weighted index, not a combined pass percentage. No composite confidence interval is available. Small differences do not establish significant superiority. Cost excludes tools, infrastructure and unreported billing dimensions; it is not measured cost across the full benchmark suite.
| Rank | Configuration | Lens score / 100 | Quality / 100 | Evidence | DeepSWE est. $ / attempt | Terminal agent | Frontier |
|---|---|---|---|---|---|---|---|
| 1 | GPT-6 Astra · xhigh | 54.39 | 72.41 | Estimated · 70% | $4.43 | Codex | Yes |
| 2 | GPT-6 Astra · max | 53.93 | 71.84 | Measured · 100% | $7.50 | Codex | No |
| 3 | GPT-6 Astra · high | 53.91 | 71.75 | Estimated · 70% | $3.92 | Codex | Yes |
| 4 | GPT-5.5 · xhigh | 53.86 | 71.75 | Estimated · 70% | $7.11 | No exact match | No |
| 5 | GPT-6 Astra · medium | 52.87 | 70.33 | Estimated · 70% | $3.08 | Codex | Yes |
Sources, weights, and coverage
Snapshot checked September 22, 2026 UTC. Weights are Lens’s coding-workflow preference, not scientifically optimal weights. Equal means within LiveBench categories prevent categories with more subtasks from gaining extra weight.
- DeepSWE v1.1 — 40% · repository engineering · mini-SWE-agent · 113 tasks
- Terminal-Bench 4.0 — 30% · terminal workflows · published model/agent pairs; agents differ
- LiveBench 2026-06-25 — 30% · coding 15%, reasoning 10%, instructions 5% · equal subtask means within each category
All reasoning levels includes measured and estimated scores; Measured only requires all five measured checks. Observed values always retain exact model and effort matches. For missing components, use the nearest measured effort of the same model, multiply its component score by target DeepSWE pass rate / anchor DeepSWE pass rate, and bound to 0–100. A zero-pass anchor is never used. Without an anchor, use the target DeepSWE score as an explicitly labeled proxy. Weights never redistribute. LiveBench rows with unspecified effort are excluded. Terminal agents vary by provider and are shown in the table. Scores describe those published systems.
Lens score = 0.75 × quality + 0.25 × (cheapest eligible DeepSWE cost ÷ this configuration’s DeepSWE cost × 100). Rank can change as the eligible cohort changes. Tokens are already reflected in cost and receive no separate bonus; no latency ranking is claimed.
LiveCodeBench’s downloaded default data ended in April 2025 and lacked current model coverage. SWE-Bench Pro was left out to avoid adding another overlapping repository benchmark with mixed harnesses. Neither contributes hidden scores.
| Configuration | DeepSWE · engineering (40%) | Terminal-Bench · terminal work (30%) | LiveBench · coding (15%) | LiveBench · reasoning (10%) | LiveBench · instructions (5%) | Quality / 100 | Coverage by weight | Status |
|---|---|---|---|---|---|---|---|---|
| GPT-6 Astra · xhigh | 74.12 | 57.88 | 81.33Estimated: Estimated from gpt-6-astra · max: 80.36 × (74.12 / 73.23), bounded to 0–100 | 93.77Estimated: Estimated from gpt-6-astra · max: 92.65 × (74.12 / 73.23), bounded to 0–100 | 76.49Estimated: Estimated from gpt-6-astra · max: 75.58 × (74.12 / 73.23), bounded to 0–100 | 72.41 | 70% | Provisional rank |
| GPT-6 Astra · maxLiveBench: gpt-6-astra-max | 73.23 | 58.18 | 80.36 | 92.65 | 75.58 | 71.84 | 100% | Measured components |
| GPT-6 Astra · high | 73.23 | 57.88 | 80.36Estimated: Estimated from gpt-6-astra · max: 80.36 × (73.23 / 73.23), bounded to 0–100 | 92.65Estimated: Estimated from gpt-6-astra · max: 92.65 × (73.23 / 73.23), bounded to 0–100 | 75.58Estimated: Estimated from gpt-6-astra · max: 75.58 × (73.23 / 73.23), bounded to 0–100 | 71.75 | 70% | Provisional rank |
| GPT-5.5 · xhighLiveBench: gpt-5.5-xhigh | 67.04 | 67.04Estimated: DeepSWE-only proxy: no same-model measured anchor available | 82.15 | 89.65 | 70.73 | 71.75 | 70% | Provisional rank |
| GPT-6 Astra · medium | 72.79 | 54.24 | 79.87Estimated: Estimated from gpt-6-astra · max: 80.36 × (72.79 / 73.23), bounded to 0–100 | 92.09Estimated: Estimated from gpt-6-astra · max: 92.65 × (72.79 / 73.23), bounded to 0–100 | 75.12Estimated: Estimated from gpt-6-astra · max: 75.58 × (72.79 / 73.23), bounded to 0–100 | 70.33 | 70% | Provisional rank |
| Opus 5 · xhigh | 73.15 | 53.94 | 80.90Estimated: Estimated from claude-opus-5 · max: 81.45 × (73.15 / 73.65), bounded to 0–100 | 90.60Estimated: Estimated from claude-opus-5 · max: 91.21 × (73.15 / 73.65), bounded to 0–100 | 63.34Estimated: Estimated from claude-opus-5 · max: 63.77 × (73.15 / 73.65), bounded to 0–100 | 69.81 | 70% | Provisional rank |
| Opus 5 · maxLiveBench: claude-opus-5-max-effort | 73.65 | 51.82 | 81.45 | 91.21 | 63.77 | 69.53 | 100% | Measured components |
| GPT-5.5 · high | 64.38 | 64.38Estimated: DeepSWE-only proxy: no same-model measured anchor available | 78.90Estimated: Estimated from gpt-5.5 · xhigh: 82.15 × (64.38 / 67.04), bounded to 0–100 | 86.10Estimated: Estimated from gpt-5.5 · xhigh: 89.65 × (64.38 / 67.04), bounded to 0–100 | 67.93Estimated: Estimated from gpt-5.5 · xhigh: 70.73 × (64.38 / 67.04), bounded to 0–100 | 68.91 | 40% | Provisional rank |
| Kimi K3 · max | 68.51 | 68.51Estimated: DeepSWE-only proxy: no same-model measured anchor available | 68.51Estimated: DeepSWE-only proxy: no same-model measured anchor available | 68.51Estimated: DeepSWE-only proxy: no same-model measured anchor available | 68.51Estimated: DeepSWE-only proxy: no same-model measured anchor available | 68.51 | 40% | Provisional rank |
| Opus 5 · high | 72.83 | 50.30 | 80.54Estimated: Estimated from claude-opus-5 · max: 81.45 × (72.83 / 73.65), bounded to 0–100 | 90.20Estimated: Estimated from claude-opus-5 · max: 91.21 × (72.83 / 73.65), bounded to 0–100 | 63.06Estimated: Estimated from claude-opus-5 · max: 63.77 × (72.83 / 73.65), bounded to 0–100 | 68.47 | 70% | Provisional rank |
| Fable 5 · xhigh | 69.91 | 44.67Estimated: Estimated from claude-fable-5 · max: 44.55 × (69.91 / 69.72), bounded to 0–100 | 86.22Estimated: Estimated from claude-fable-5 · max: 85.99 × (69.91 / 69.72), bounded to 0–100 | 89.89Estimated: Estimated from claude-fable-5 · max: 89.65 × (69.91 / 69.72), bounded to 0–100 | 75.97Estimated: Estimated from claude-fable-5 · max: 75.77 × (69.91 / 69.72), bounded to 0–100 | 67.09 | 40% | Provisional rank |
| Fable 5 · maxLiveBench: claude-fable-5-max-effort | 69.72 | 44.55 | 85.99 | 89.65 | 75.77 | 66.91 | 100% | Measured components |
| Fable 5 · high | 68.60 | 43.83Estimated: Estimated from claude-fable-5 · max: 44.55 × (68.60 / 69.72), bounded to 0–100 | 84.61Estimated: Estimated from claude-fable-5 · max: 85.99 × (68.60 / 69.72), bounded to 0–100 | 88.21Estimated: Estimated from claude-fable-5 · max: 89.65 × (68.60 / 69.72), bounded to 0–100 | 74.55Estimated: Estimated from claude-fable-5 · max: 75.77 × (68.60 / 69.72), bounded to 0–100 | 65.83 | 40% | Provisional rank |
| GPT-5.6 Sol · maxLiveBench: gpt-5.6-sol-max | 72.67 | 37.27 | 83.94 | 91.65 | 71.85 | 65.60 | 100% | Measured components |
| GPT-6 Astra · low | 67.04 | 50.61 | 73.56Estimated: Estimated from gpt-6-astra · max: 80.36 × (67.04 / 73.23), bounded to 0–100 | 84.82Estimated: Estimated from gpt-6-astra · max: 92.65 × (67.04 / 73.23), bounded to 0–100 | 69.19Estimated: Estimated from gpt-6-astra · max: 75.58 × (67.04 / 73.23), bounded to 0–100 | 64.97 | 70% | Provisional rank |
| Opus 5 · medium | 68.90 | 44.85 | 76.20Estimated: Estimated from claude-opus-5 · max: 81.45 × (68.90 / 73.65), bounded to 0–100 | 85.34Estimated: Estimated from claude-opus-5 · max: 91.21 × (68.90 / 73.65), bounded to 0–100 | 59.66Estimated: Estimated from claude-opus-5 · max: 63.77 × (68.90 / 73.65), bounded to 0–100 | 63.96 | 70% | Provisional rank |
| GLM 5.3 Flash · max | 63.39 | 63.39Estimated: DeepSWE-only proxy: no same-model measured anchor available | 63.39Estimated: DeepSWE-only proxy: no same-model measured anchor available | 63.39Estimated: DeepSWE-only proxy: no same-model measured anchor available | 63.39Estimated: DeepSWE-only proxy: no same-model measured anchor available | 63.39 | 40% | Provisional rank |
| GPT-5.6 Sol · xhigh | 70.73 | 36.28Estimated: Estimated from gpt-5.6-sol · max: 37.27 × (70.73 / 72.67), bounded to 0–100 | 81.71Estimated: Estimated from gpt-5.6-sol · max: 83.94 × (70.73 / 72.67), bounded to 0–100 | 89.21Estimated: Estimated from gpt-5.6-sol · max: 91.65 × (70.73 / 72.67), bounded to 0–100 | 69.93Estimated: Estimated from gpt-5.6-sol · max: 71.85 × (70.73 / 72.67), bounded to 0–100 | 63.85 | 40% | Provisional rank |
| DeepSeek V4 Pro · max | 62.83 | 62.83Estimated: DeepSWE-only proxy: no same-model measured anchor available | 62.83Estimated: DeepSWE-only proxy: no same-model measured anchor available | 62.83Estimated: DeepSWE-only proxy: no same-model measured anchor available | 62.83Estimated: DeepSWE-only proxy: no same-model measured anchor available | 62.83 | 40% | Provisional rank |
| Muse Spark 1.2 · xhighLiveBench: muse-spark-1.2-xhigh | 54.87 | 54.87Estimated: DeepSWE-only proxy: no same-model measured anchor available | 77.54 | 90.00 | 74.33 | 62.76 | 70% | Provisional rank |
| GPT-5.6 Sol · high | 69.40 | 35.60Estimated: Estimated from gpt-5.6-sol · max: 37.27 × (69.40 / 72.67), bounded to 0–100 | 80.17Estimated: Estimated from gpt-5.6-sol · max: 83.94 × (69.40 / 72.67), bounded to 0–100 | 87.54Estimated: Estimated from gpt-5.6-sol · max: 91.65 × (69.40 / 72.67), bounded to 0–100 | 68.62Estimated: Estimated from gpt-5.6-sol · max: 71.85 × (69.40 / 72.67), bounded to 0–100 | 62.65 | 40% | Provisional rank |
| Fable 5 · medium | 65.37 | 41.77Estimated: Estimated from claude-fable-5 · max: 44.55 × (65.37 / 69.72), bounded to 0–100 | 80.62Estimated: Estimated from claude-fable-5 · max: 85.99 × (65.37 / 69.72), bounded to 0–100 | 84.05Estimated: Estimated from claude-fable-5 · max: 89.65 × (65.37 / 69.72), bounded to 0–100 | 71.04Estimated: Estimated from claude-fable-5 · max: 75.77 × (65.37 / 69.72), bounded to 0–100 | 62.73 | 40% | Provisional rank |
| Muse Spark 1.1 · xhighLiveBench: muse-spark-1.1-xhigh | 53.32 | 53.32Estimated: DeepSWE-only proxy: no same-model measured anchor available | 77.16 | 87.73 | 69.64 | 61.15 | 70% | Provisional rank |
| GLM 5.3 · max | 68.96 | 41.82 | 68.96Estimated: DeepSWE-only proxy: no same-model measured anchor available | 68.96Estimated: DeepSWE-only proxy: no same-model measured anchor available | 68.96Estimated: DeepSWE-only proxy: no same-model measured anchor available | 60.82 | 70% | Provisional rank |
| GPT-5.4 · xhighLiveBench: gpt-5.4-xhigh | 51.77 | 51.77Estimated: DeepSWE-only proxy: no same-model measured anchor available | 77.54 | 88.12 | 70.22 | 60.19 | 70% | Provisional rank |
| Gemini 3.8 Flash · highLiveBench: gemini-3.8-flash-high | 73.83 | 19.09 | 72.49 | 89.29 | 81.41 | 59.13 | 100% | Measured components |
| GPT-5.6 Terra · maxLiveBench: gpt-5.6-terra-max | 69.62 | 21.52 | 78.25 | 90.63 | 64.62 | 58.34 | 100% | Measured components |
| GPT-5.5 · medium | 53.98 | 53.98Estimated: DeepSWE-only proxy: no same-model measured anchor available | 66.15Estimated: Estimated from gpt-5.5 · xhigh: 82.15 × (53.98 / 67.04), bounded to 0–100 | 72.20Estimated: Estimated from gpt-5.5 · xhigh: 89.65 × (53.98 / 67.04), bounded to 0–100 | 56.96Estimated: Estimated from gpt-5.5 · xhigh: 70.73 × (53.98 / 67.04), bounded to 0–100 | 57.78 | 40% | Provisional rank |
| Fable 5 · low | 59.58 | 38.07Estimated: Estimated from claude-fable-5 · max: 44.55 × (59.58 / 69.72), bounded to 0–100 | 73.49Estimated: Estimated from claude-fable-5 · max: 85.99 × (59.58 / 69.72), bounded to 0–100 | 76.61Estimated: Estimated from claude-fable-5 · max: 89.65 × (59.58 / 69.72), bounded to 0–100 | 64.75Estimated: Estimated from claude-fable-5 · max: 75.77 × (59.58 / 69.72), bounded to 0–100 | 57.18 | 40% | Provisional rank |
| Gemini 3.8 Flash · medium | 71.02 | 18.36Estimated: Estimated from gemini-3.8-flash · high: 19.09 × (71.02 / 73.83), bounded to 0–100 | 69.73Estimated: Estimated from gemini-3.8-flash · high: 72.49 × (71.02 / 73.83), bounded to 0–100 | 85.90Estimated: Estimated from gemini-3.8-flash · high: 89.29 × (71.02 / 73.83), bounded to 0–100 | 78.32Estimated: Estimated from gemini-3.8-flash · high: 81.41 × (71.02 / 73.83), bounded to 0–100 | 56.88 | 40% | Provisional rank |
| GPT-5.6 Luna · maxLiveBench: gpt-5.6-luna-max | 67.19 | 17.27 | 82.91 | 85.64 | 60.12 | 56.06 | 100% | Measured components |
| Gemini 3.6 Flash · highLiveBench: gemini-3.6-flash-high | 46.68 | 46.68Estimated: DeepSWE-only proxy: no same-model measured anchor available | 77.86 | 85.15 | 75.37 | 56.64 | 70% | Provisional rank |
| Opus 4.8 · maxLiveBench: claude-opus-4-8-max-effort | 58.97 | 23.64 | 81.83 | 89.19 | 72.03 | 55.48 | 100% | Measured components |
| GPT-5.6 Sol · medium | 61.06 | 31.32Estimated: Estimated from gpt-5.6-sol · max: 37.27 × (61.06 / 72.67), bounded to 0–100 | 70.54Estimated: Estimated from gpt-5.6-sol · max: 83.94 × (61.06 / 72.67), bounded to 0–100 | 77.02Estimated: Estimated from gpt-5.6-sol · max: 91.65 × (61.06 / 72.67), bounded to 0–100 | 60.37Estimated: Estimated from gpt-5.6-sol · max: 71.85 × (61.06 / 72.67), bounded to 0–100 | 55.12 | 40% | Provisional rank |
| DeepSeek V4 Flash · max | 53.32 | 53.32Estimated: DeepSWE-only proxy: no same-model measured anchor available | 53.32Estimated: DeepSWE-only proxy: no same-model measured anchor available | 53.32Estimated: DeepSWE-only proxy: no same-model measured anchor available | 53.32Estimated: DeepSWE-only proxy: no same-model measured anchor available | 53.32 | 40% | Provisional rank |
| Gemini 3.7 Flash · medium | 65.49 | 11.25Estimated: Estimated from gemini-3.7-flash · high: 11.21 × (65.49 / 65.27), bounded to 0–100 | 79.16Estimated: Estimated from gemini-3.7-flash · high: 78.89 × (65.49 / 65.27), bounded to 0–100 | 88.10Estimated: Estimated from gemini-3.7-flash · high: 87.80 × (65.49 / 65.27), bounded to 0–100 | 80.20Estimated: Estimated from gemini-3.7-flash · high: 79.93 × (65.49 / 65.27), bounded to 0–100 | 54.26 | 40% | Provisional rank |
| Gemini 3.7 Flash · highLiveBench: gemini-3.7-flash-high | 65.27 | 11.21 | 78.89 | 87.80 | 79.93 | 54.08 | 100% | Measured components |
| Grok 4.6 · medium | 67.48 | 21.01Estimated: Estimated from grok-4.6 · high: 20.30 × (67.48 / 65.19), bounded to 0–100 | 67.48Estimated: DeepSWE-only proxy: no same-model measured anchor available | 67.48Estimated: DeepSWE-only proxy: no same-model measured anchor available | 67.48Estimated: DeepSWE-only proxy: no same-model measured anchor available | 53.54 | 40% | Provisional rank |
| Opus 5 · low | 58.13 | 34.85 | 64.28Estimated: Estimated from claude-opus-5 · max: 81.45 × (58.13 / 73.65), bounded to 0–100 | 71.99Estimated: Estimated from claude-opus-5 · max: 91.21 × (58.13 / 73.65), bounded to 0–100 | 50.33Estimated: Estimated from claude-opus-5 · max: 63.77 × (58.13 / 73.65), bounded to 0–100 | 53.06 | 70% | Provisional rank |
| Grok 4.6 · xhigh | 66.74 | 20.78Estimated: Estimated from grok-4.6 · high: 20.30 × (66.74 / 65.19), bounded to 0–100 | 66.74Estimated: DeepSWE-only proxy: no same-model measured anchor available | 66.74Estimated: DeepSWE-only proxy: no same-model measured anchor available | 66.74Estimated: DeepSWE-only proxy: no same-model measured anchor available | 52.95 | 40% | Provisional rank |
| Grok 4.6 · high | 65.19 | 20.30 | 65.19Estimated: DeepSWE-only proxy: no same-model measured anchor available | 65.19Estimated: DeepSWE-only proxy: no same-model measured anchor available | 65.19Estimated: DeepSWE-only proxy: no same-model measured anchor available | 51.72 | 70% | Provisional rank |
| Sonnet 5 · max | 53.85 | 12.42 | 87.47Estimated: Estimated from claude-sonnet-5 · xhigh: 80.68 × (53.85 / 49.67), bounded to 0–100 | 96.15Estimated: Estimated from claude-sonnet-5 · xhigh: 88.69 × (53.85 / 49.67), bounded to 0–100 | 69.23Estimated: Estimated from claude-sonnet-5 · xhigh: 63.86 × (53.85 / 49.67), bounded to 0–100 | 51.46 | 70% | Provisional rank |
| Opus 4.8 · xhigh | 54.36 | 21.79Estimated: Estimated from claude-opus-4-8 · max: 23.64 × (54.36 / 58.97), bounded to 0–100 | 75.43Estimated: Estimated from claude-opus-4-8 · max: 81.83 × (54.36 / 58.97), bounded to 0–100 | 82.22Estimated: Estimated from claude-opus-4-8 · max: 89.19 × (54.36 / 58.97), bounded to 0–100 | 66.40Estimated: Estimated from claude-opus-4-8 · max: 72.03 × (54.36 / 58.97), bounded to 0–100 | 51.14 | 40% | Provisional rank |
| GPT-5.6 Terra · xhigh | 60.18 | 18.60Estimated: Estimated from gpt-5.6-terra · max: 21.52 × (60.18 / 69.62), bounded to 0–100 | 67.63Estimated: Estimated from gpt-5.6-terra · max: 78.25 × (60.18 / 69.62), bounded to 0–100 | 78.34Estimated: Estimated from gpt-5.6-terra · max: 90.63 × (60.18 / 69.62), bounded to 0–100 | 55.85Estimated: Estimated from gpt-5.6-terra · max: 64.62 × (60.18 / 69.62), bounded to 0–100 | 50.42 | 40% | Provisional rank |
| Gemini 3.5 Flash · highLiveBench: gemini-3.5-flash-high | 36.06 | 36.06Estimated: DeepSWE-only proxy: no same-model measured anchor available | 78.18 | 82.00 | 75.60 | 48.95 | 70% | Provisional rank |
| GPT-5.6 Luna · xhigh | 56.86 | 14.61Estimated: Estimated from gpt-5.6-luna · max: 17.27 × (56.86 / 67.19), bounded to 0–100 | 70.17Estimated: Estimated from gpt-5.6-luna · max: 82.91 × (56.86 / 67.19), bounded to 0–100 | 72.48Estimated: Estimated from gpt-5.6-luna · max: 85.64 × (56.86 / 67.19), bounded to 0–100 | 50.88Estimated: Estimated from gpt-5.6-luna · max: 60.12 × (56.86 / 67.19), bounded to 0–100 | 47.44 | 40% | Provisional rank |
| Opus 4.8 · high | 51.77 | 20.75Estimated: Estimated from claude-opus-4-8 · max: 23.64 × (51.77 / 58.97), bounded to 0–100 | 71.83Estimated: Estimated from claude-opus-4-8 · max: 81.83 × (51.77 / 58.97), bounded to 0–100 | 78.30Estimated: Estimated from claude-opus-4-8 · max: 89.19 × (51.77 / 58.97), bounded to 0–100 | 63.23Estimated: Estimated from claude-opus-4-8 · max: 72.03 × (51.77 / 58.97), bounded to 0–100 | 48.70 | 40% | Provisional rank |
| Sonnet 5 · xhighLiveBench: claude-sonnet-5-xhigh-effort | 49.67 | 11.46Estimated: Estimated from claude-sonnet-5 · max: 12.42 × (49.67 / 53.85), bounded to 0–100 | 80.68 | 88.69 | 63.86 | 47.47 | 70% | Provisional rank |
| Sonnet 5 · high | 48.23 | 11.12Estimated: Estimated from claude-sonnet-5 · max: 12.42 × (48.23 / 53.85), bounded to 0–100 | 78.35Estimated: Estimated from claude-sonnet-5 · xhigh: 80.68 × (48.23 / 49.67), bounded to 0–100 | 86.13Estimated: Estimated from claude-sonnet-5 · xhigh: 88.69 × (48.23 / 49.67), bounded to 0–100 | 62.01Estimated: Estimated from claude-sonnet-5 · xhigh: 63.86 × (48.23 / 49.67), bounded to 0–100 | 46.09 | 40% | Provisional rank |
| Opus 4.8 · medium | 48.67 | 19.51Estimated: Estimated from claude-opus-4-8 · max: 23.64 × (48.67 / 58.97), bounded to 0–100 | 67.53Estimated: Estimated from claude-opus-4-8 · max: 81.83 × (48.67 / 58.97), bounded to 0–100 | 73.61Estimated: Estimated from claude-opus-4-8 · max: 89.19 × (48.67 / 58.97), bounded to 0–100 | 59.45Estimated: Estimated from claude-opus-4-8 · max: 72.03 × (48.67 / 58.97), bounded to 0–100 | 45.79 | 40% | Provisional rank |
| GPT-5.6 Terra · high | 53.76 | 16.62Estimated: Estimated from gpt-5.6-terra · max: 21.52 × (53.76 / 69.62), bounded to 0–100 | 60.42Estimated: Estimated from gpt-5.6-terra · max: 78.25 × (53.76 / 69.62), bounded to 0–100 | 69.99Estimated: Estimated from gpt-5.6-terra · max: 90.63 × (53.76 / 69.62), bounded to 0–100 | 49.90Estimated: Estimated from gpt-5.6-terra · max: 64.62 × (53.76 / 69.62), bounded to 0–100 | 45.05 | 40% | Provisional rank |
| Gemini 3.7 Flash · low | 53.76 | 9.23Estimated: Estimated from gemini-3.7-flash · high: 11.21 × (53.76 / 65.27), bounded to 0–100 | 64.98Estimated: Estimated from gemini-3.7-flash · high: 78.89 × (53.76 / 65.27), bounded to 0–100 | 72.32Estimated: Estimated from gemini-3.7-flash · high: 87.80 × (53.76 / 65.27), bounded to 0–100 | 65.84Estimated: Estimated from gemini-3.7-flash · high: 79.93 × (53.76 / 65.27), bounded to 0–100 | 44.55 | 40% | Provisional rank |
| GLM 5.2 · max | 43.78 | 43.78Estimated: DeepSWE-only proxy: no same-model measured anchor available | 43.78Estimated: DeepSWE-only proxy: no same-model measured anchor available | 43.78Estimated: DeepSWE-only proxy: no same-model measured anchor available | 43.78Estimated: DeepSWE-only proxy: no same-model measured anchor available | 43.78 | 40% | Provisional rank |
| Kimi K2.7 CodeLiveBench: kimi-k2.7-code | 30.53 | 30.53Estimated: DeepSWE-only proxy: no same-model measured anchor available | 73.96 | 82.81 | 56.29 | 43.56 | 70% | Provisional rank |
| Grok 4.5 · high | 53.76 | 12.42 | 53.76Estimated: DeepSWE-only proxy: no same-model measured anchor available | 53.76Estimated: DeepSWE-only proxy: no same-model measured anchor available | 53.76Estimated: DeepSWE-only proxy: no same-model measured anchor available | 41.36 | 70% | Provisional rank |
| GPT-5.6 Sol · low | 45.35 | 23.26Estimated: Estimated from gpt-5.6-sol · max: 37.27 × (45.35 / 72.67), bounded to 0–100 | 52.39Estimated: Estimated from gpt-5.6-sol · max: 83.94 × (45.35 / 72.67), bounded to 0–100 | 57.20Estimated: Estimated from gpt-5.6-sol · max: 91.65 × (45.35 / 72.67), bounded to 0–100 | 44.84Estimated: Estimated from gpt-5.6-sol · max: 71.85 × (45.35 / 72.67), bounded to 0–100 | 40.94 | 40% | Provisional rank |
| GPT-5.6 Luna · high | 44.25 | 11.37Estimated: Estimated from gpt-5.6-luna · max: 17.27 × (44.25 / 67.19), bounded to 0–100 | 54.61Estimated: Estimated from gpt-5.6-luna · max: 82.91 × (44.25 / 67.19), bounded to 0–100 | 56.40Estimated: Estimated from gpt-5.6-luna · max: 85.64 × (44.25 / 67.19), bounded to 0–100 | 39.59Estimated: Estimated from gpt-5.6-luna · max: 60.12 × (44.25 / 67.19), bounded to 0–100 | 36.92 | 40% | Provisional rank |
| Opus 4.8 · low | 40.80 | 16.35Estimated: Estimated from claude-opus-4-8 · max: 23.64 × (40.80 / 58.97), bounded to 0–100 | 56.61Estimated: Estimated from claude-opus-4-8 · max: 81.83 × (40.80 / 58.97), bounded to 0–100 | 61.70Estimated: Estimated from claude-opus-4-8 · max: 89.19 × (40.80 / 58.97), bounded to 0–100 | 49.83Estimated: Estimated from claude-opus-4-8 · max: 72.03 × (40.80 / 58.97), bounded to 0–100 | 38.38 | 40% | Provisional rank |
| Sonnet 5 · medium | 39.78 | 9.18Estimated: Estimated from claude-sonnet-5 · max: 12.42 × (39.78 / 53.85), bounded to 0–100 | 64.62Estimated: Estimated from claude-sonnet-5 · xhigh: 80.68 × (39.78 / 49.67), bounded to 0–100 | 71.03Estimated: Estimated from claude-sonnet-5 · xhigh: 88.69 × (39.78 / 49.67), bounded to 0–100 | 51.14Estimated: Estimated from claude-sonnet-5 · xhigh: 63.86 × (39.78 / 49.67), bounded to 0–100 | 38.02 | 40% | Provisional rank |
| GLM 5.2 · high | 36.28 | 36.28Estimated: DeepSWE-only proxy: no same-model measured anchor available | 36.28Estimated: DeepSWE-only proxy: no same-model measured anchor available | 36.28Estimated: DeepSWE-only proxy: no same-model measured anchor available | 36.28Estimated: DeepSWE-only proxy: no same-model measured anchor available | 36.28 | 40% | Provisional rank |
| GPT-5.6 Luna · low | 1.55 | 0.40Estimated: Estimated from gpt-5.6-luna · max: 17.27 × (1.55 / 67.19), bounded to 0–100 | 1.91Estimated: Estimated from gpt-5.6-luna · max: 82.91 × (1.55 / 67.19), bounded to 0–100 | 1.97Estimated: Estimated from gpt-5.6-luna · max: 85.64 × (1.55 / 67.19), bounded to 0–100 | 1.39Estimated: Estimated from gpt-5.6-luna · max: 60.12 × (1.55 / 67.19), bounded to 0–100 | 1.29 | 40% | Provisional rank |
| Grok 4.6 · low | 41.65 | 12.97Estimated: Estimated from grok-4.6 · high: 20.30 × (41.65 / 65.19), bounded to 0–100 | 41.65Estimated: DeepSWE-only proxy: no same-model measured anchor available | 41.65Estimated: DeepSWE-only proxy: no same-model measured anchor available | 41.65Estimated: DeepSWE-only proxy: no same-model measured anchor available | 33.04 | 40% | Provisional rank |
| Gemini 3.1 Pro Preview · highLiveBench: gemini-3.1-pro-preview-high | 11.73 | 11.73Estimated: DeepSWE-only proxy: no same-model measured anchor available | 76.45 | 84.00 | 79.10 | 32.03 | 70% | Provisional rank |
| GPT-5.6 Terra · medium | 35.11 | 10.85Estimated: Estimated from gpt-5.6-terra · max: 21.52 × (35.11 / 69.62), bounded to 0–100 | 39.46Estimated: Estimated from gpt-5.6-terra · max: 78.25 × (35.11 / 69.62), bounded to 0–100 | 45.71Estimated: Estimated from gpt-5.6-terra · max: 90.63 × (35.11 / 69.62), bounded to 0–100 | 32.59Estimated: Estimated from gpt-5.6-terra · max: 64.62 × (35.11 / 69.62), bounded to 0–100 | 29.42 | 40% | Provisional rank |
| Sonnet 4.6 · high | 29.93 | 29.93Estimated: DeepSWE-only proxy: no same-model measured anchor available | 29.93Estimated: DeepSWE-only proxy: no same-model measured anchor available | 29.93Estimated: DeepSWE-only proxy: no same-model measured anchor available | 29.93Estimated: DeepSWE-only proxy: no same-model measured anchor available | 29.93 | 40% | Provisional rank |
| Sonnet 5 · low | 30.51 | 7.04Estimated: Estimated from claude-sonnet-5 · max: 12.42 × (30.51 / 53.85), bounded to 0–100 | 49.56Estimated: Estimated from claude-sonnet-5 · xhigh: 80.68 × (30.51 / 49.67), bounded to 0–100 | 54.49Estimated: Estimated from claude-sonnet-5 · xhigh: 88.69 × (30.51 / 49.67), bounded to 0–100 | 39.23Estimated: Estimated from claude-sonnet-5 · xhigh: 63.86 × (30.51 / 49.67), bounded to 0–100 | 29.16 | 40% | Provisional rank |
| GPT-5.5 · low | 26.99 | 26.99Estimated: DeepSWE-only proxy: no same-model measured anchor available | 33.08Estimated: Estimated from gpt-5.5 · xhigh: 82.15 × (26.99 / 67.04), bounded to 0–100 | 36.10Estimated: Estimated from gpt-5.5 · xhigh: 89.65 × (26.99 / 67.04), bounded to 0–100 | 28.48Estimated: Estimated from gpt-5.5 · xhigh: 70.73 × (26.99 / 67.04), bounded to 0–100 | 28.89 | 40% | Provisional rank |
| GPT-5.6 Terra · low | 24.05 | 7.43Estimated: Estimated from gpt-5.6-terra · max: 21.52 × (24.05 / 69.62), bounded to 0–100 | 27.03Estimated: Estimated from gpt-5.6-terra · max: 78.25 × (24.05 / 69.62), bounded to 0–100 | 31.31Estimated: Estimated from gpt-5.6-terra · max: 90.63 × (24.05 / 69.62), bounded to 0–100 | 22.32Estimated: Estimated from gpt-5.6-terra · max: 64.62 × (24.05 / 69.62), bounded to 0–100 | 20.15 | 40% | Provisional rank |
| GPT-5.6 Luna · medium | 11.28 | 2.90Estimated: Estimated from gpt-5.6-luna · max: 17.27 × (11.28 / 67.19), bounded to 0–100 | 13.92Estimated: Estimated from gpt-5.6-luna · max: 82.91 × (11.28 / 67.19), bounded to 0–100 | 14.38Estimated: Estimated from gpt-5.6-luna · max: 85.64 × (11.28 / 67.19), bounded to 0–100 | 10.10Estimated: Estimated from gpt-5.6-luna · max: 60.12 × (11.28 / 67.19), bounded to 0–100 | 9.42 | 40% | Provisional rank |