Flame-VLM-Code (Flame-VLM-Code)

A vision-language coding benchmark for generating correct code from visual and multimodal inputs.

Top Models on Flame-VLM-Code — March 2026

As of March 2026, Claude Opus 4.6 leads the Flame-VLM-Code leaderboard with 98.8% , followed by GLM-5V-Turbo (93.8%) and Kimi K2.5 (88.8%).

3 modelsMultimodal & GroundedCurrentDisplay onlyUpdated April 1, 2026

According to BenchLM.ai, Claude Opus 4.6 leads the Flame-VLM-Code benchmark with a score of 98.8%, followed by GLM-5V-Turbo (93.8%) and Kimi K2.5 (88.8%). The scores show moderate spread, with meaningful differences between the top tier and mid-tier models.

3 models have been evaluated on Flame-VLM-Code. The benchmark falls in the Multimodal & Grounded category. This category carries a 12% weight in BenchLM.ai's overall scoring system. Flame-VLM-Code is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.

About Flame-VLM-Code

Year

2026

Tasks

Multimodal coding tasks

Format

Vision-language code generation

Difficulty

Multimodal coding

BenchLM tracks Flame-VLM-Code as a display-only multimodal coding benchmark reference.

GLM-5V-Turbo

BenchLM freshness & provenance

Version

Flame-VLM-Code 2026

Refresh cadence

Quarterly

Staleness state

Current

Question availability

Public benchmark set

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.

Leaderboard (3 models)

#1Claude Opus 4.6
98.8%
#2GLM-5V-Turbo
93.8%
#3Kimi K2.5
88.8%

FAQ

What does Flame-VLM-Code measure?

A vision-language coding benchmark for generating correct code from visual and multimodal inputs.

Which model scores highest on Flame-VLM-Code?

Claude Opus 4.6 by Anthropic currently leads with a score of 98.8% on Flame-VLM-Code.

How many models are evaluated on Flame-VLM-Code?

3 AI models have been evaluated on Flame-VLM-Code on BenchLM.

Last updated: April 1, 2026 · BenchLM version Flame-VLM-Code 2026

Weekly LLM Benchmark Digest

Get notified when new models drop, benchmark scores change, or the leaderboard shifts. One email per week.

Free. No spam. Unsubscribe anytime. We only store derived location metadata for consent routing.