Skip to main content
BenchLM
Data

BridgeBench

We show this table for reference; we do not rank on it.

Data verified

BridgeBench publishes comparative model ratings across nine software-building axes, including reasoning, front end, back end, security, speed, and cost. Its scoring methodology is proprietary.

Overall editorial rating on BridgeBench — October 19, 2026

We mirror the published overall editorial rating view for BridgeBench. Claude Opus 5.5 leads the public snapshot at 760, followed by GPT-6 Astra (728) and Claude Fable 5.1 (689). We do not use these results to rank models overall.

1Closed

Claude Opus 5.5

Anthropic

760
Overall 86.62Context 1M
2Closed

GPT-6 Astra

OpenAI

728
Overall 85.65Context 1.05M
3Closed

Claude Fable 5.1

Anthropic

689
Overall 79.21Context 1M

30 modelsCodingCurrentDisplay onlyUpdated October 19, 2026

Overall editorial rating table (30 models)

Overall editorial rating results for BridgeBench
RankModel / configurationScoreParameters (B)Open / closed
1Claude Opus 5.5Anthropic
760
Not reportedClosed
2GPT-6 AstraOpenAI
728
Not reportedClosed
3Claude Fable 5.1Anthropic
689
Not reportedClosed
4Claude Sonnet 5.5Anthropic
647
Not reportedClosed
5GPT-6.1 SolOpenAI
646
Not reportedClosed
6GPT-6 SolOpenAI
643
Not reportedClosed
7Claude Fable 5Anthropic
628
Not reportedClosed
8Gemini 4 ArgonGoogle
623
Not reportedClosed
9Muse WonderMeta
611
Not reportedClosed
10Muse Spark 1.3Meta
606
Not reportedClosed
11Claude Haiku 5.5Anthropic
601
Not reportedClosed
12GPT-5.6 SolOpenAI
600
Not reportedClosed
13Gemini 3.8 FlashGoogle
584
Not reportedClosed
14Step 5 PreviewStepFun
572
Not reportedOpen
15MiMo-V2.6-ProXiaomi
568
Not reportedOpen
16DeepSeek V4.1 FlashDeepSeek
564
Not reportedUnknown
17Grok 4.7xAI
540
Not reportedClosed
18Grok 4.6xAI
536
Not reportedClosed
19Qwen3.8 MaxAlibaba
532
Not reportedOpen
20Qwen3.8-Flash-NextAlibaba
522
Not reportedOpen
21GPT-6 LunaOpenAI
513
Not reportedClosed
22Claude Opus 5Anthropic
512
Not reportedClosed
23GPT-5.6 TerraOpenAI
511
Not reportedClosed
24Kimi K3Moonshot AI
501
Not reportedOpen
25Claude Sonnet 5Anthropic
501
Not reportedClosed
26GPT-5.6 LunaOpenAI
476
Not reportedClosed
27Qwen3.8 2.4T A95BAlibaba
468
Not reportedUnknown
28GLM 5.3Z.ai
461
Not reportedUnknown
29DeepSeek V4 ProDeepSeek
460
Not reportedUnknown
30GLM 5.3 FlashZ.ai
450
Not reportedUnknown

How to read this leaderboard

Use these comparative assessments as one publisher's view of software-building tasks. Its undisclosed scoring procedure does not establish performance on a shared, reproducible test suite.

Operator receipt: 30 sourced rows are currently displayable on this page; the leading published row is Claude Opus 5.5 at 760.

Honest limit: BridgeBench does not disclose its proprietary benchmarking methodology. Small rating gaps may not predict performance on a specific project. These ratings do not enter BenchLM overall or category rankings.

BridgeBench leaderboardBridgeBench scoring method

How we show BridgeBench

We captured 30 overall ratings from the public BridgeBench leaderboard on October 19, 2026. BridgeBench publishes nine axes, including reasoning, front end, back end, security, speed, and cost. Its current guide does not disclose how it calculates the overall rating.

These are BridgeBench comparative assessments. They are not head-to-head match records or pass rates from a shared test suite. The source keeps its prompts, task sets, and detailed rubrics private.

We show this table for reference only. Its ratings do not enter our overall or category rankings.

Snapshot

30 model ratings9 capability axesEditorial ratingsDisplay only
BridgeBench leaderboardHow BridgeBench scores models

The published BridgeBench snapshot places Claude Opus 5.5 first at 760. The third row is 71 score units behind. The broader top-10 range is 154 score units, so the table still separates the published systems.

30 models have been evaluated on BridgeBench. The benchmark falls in the Coding category. We keep external benchmark mirrors separate from the weighted global scoring system, so these results remain source-specific evidence. BridgeBench is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.

About BridgeBench

Year

2026

Tasks

No public task count

Format

Publisher comparative rating

Difficulty

Software-building assessments

BridgeBench reports comparative assessments rather than success percentages. Its current guide does not disclose how it calculates the overall rating. This table mirrors published overall ratings as display-only context.

Public benchmark source

Freshness and provenance

Version

September 2026 editorial ratings

Refresh cadence

Manual snapshot

Staleness state

Current

Question availability

Prompts, task sets, and detailed rubrics private

CurrentDisplay only

BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.

Questions

What does BridgeBench measure?

BridgeBench publishes comparative model ratings across nine software-building axes, including reasoning, front end, back end, security, speed, and cost. Its scoring methodology is proprietary.

Which model leads the published BridgeBench snapshot?

Claude Opus 5.5 currently leads the published BridgeBench snapshot with 760 overall editorial rating. BenchLM shows this benchmark for display only and does not use it in overall rankings.

How many models are evaluated on BridgeBench?

The October 19, 2026 snapshot contains 30 AI models.

Last updated: October 19, 2026 · mirrored from published editorial ratings

Know when it’s worth switching models

The model to choose, the cheaper alternative, and the release we would wait on.

Read a sample issue

Join 7,000+ subscribers.

One email each week. Unsubscribe anytime.