GPT-56T 861 —
MUSE-SPK 837 +0.2%
GPT-56SC 790 -4.6%
GLM-5 781 -0.4%
CL-OP55X 780 -5.1%
GROK-46H 780 -5.1%
QWEN-38X 748 -9.2%
GPT-6A 743 -9.4%
KIMI-K3X 742 —
CL-FAB5H 698 -6.1%
CL-OP5H 675 -6.2%
GEM-38FH 672 -0.7%
CL-OP5X 670 -5.5%
CL-OP55H 668 —
CL-OP46H 657 -5.9%
CL-OP47H 648 -6.1%
GPT-56S 618 -0.6%
GEM-37FH 610 -7.2%
GEM-36FH 593 —
CL-OP48H 588 —
CL-OP47 581 -0.2%
GEM-35FH 580 —
GPT-55H 541 -7%
INKL 531 —
GEM-31P 512 -0.2%
CL-OP46 498 +0.4%
GEM-3P 498 -0.2%
CL-OP48 492 +0.4%
GPT-52 464 —
GPT-55 423 —
GPT-56T 861 —
MUSE-SPK 837 +0.2%
GPT-56SC 790 -4.6%
GLM-5 781 -0.4%
CL-OP55X 780 -5.1%
GROK-46H 780 -5.1%
QWEN-38X 748 -9.2%
GPT-6A 743 -9.4%
KIMI-K3X 742 —
CL-FAB5H 698 -6.1%
CL-OP5H 675 -6.2%
GEM-38FH 672 -0.7%
CL-OP5X 670 -5.5%
CL-OP55H 668 —
CL-OP46H 657 -5.9%
CL-OP47H 648 -6.1%
GPT-56S 618 -0.6%
GEM-37FH 610 -7.2%
GEM-36FH 593 —
CL-OP48H 588 —
CL-OP47 581 -0.2%
GEM-35FH 580 —
GPT-55H 541 -7%
INKL 531 —
GEM-31P 512 -0.2%
CL-OP46 498 +0.4%
GEM-3P 498 -0.2%
CL-OP48 492 +0.4%
GPT-52 464 —
GPT-55 423 —
← Back to feed

Claude Opus 4.7 Hits 1505 Arena ELO Tied With Gemini 3.1 Pro — Three Days After Joining the Leaderboard

Claude Opus 4.7 was added to the Chatbot Arena leaderboard on April 17. By April 18 — one day after being listed — it had already accumulated enough votes to settle at 1505 ELO in the Thinking configuration, tied with Gemini 3.1 Pro Preview for the highest score on the general leaderboard.

Claude Opus 4.7 standard (without extended thinking) landed at 1503, tied with Claude Opus 4.6 Thinking. The full top of the leaderboard as of April 18:

ModelArena ELO
Claude Opus 4.7 Thinking1505
Gemini 3.1 Pro Preview1505
Claude Opus 4.71503
Claude Opus 4.6 Thinking1503
Grok 4.201496
GPT-5.4 High1495
Gemini 3 Pro1492
Claude Opus 4.61490

Coding Arena: Claude Dominates

The Coding Arena paints a sharper picture. Claude Opus 4.7 Thinking leads at 1565 Coding ELO, followed by Claude Opus 4.6 Thinking at 1545, Claude Opus 4.7 standard at 1554, and GPT-5.4 High at 1538. Gemini 3.1 Pro sits at 1531 in the Coding Arena.

Claude holds four of the top five coding slots. GPT-5.4 High is the only non-Anthropic model in the top tier.

Context: A Fast Climb

The speed of the settlement is notable. Opus 4.6 climbed from 1398 ELO in February to 1504 by early April — a two-month trajectory driven by sustained voting. Opus 4.7, benefiting from its predecessor’s reputation and significant launch coverage, appears to have entered with higher initial vote velocity. One day in, it matched the ceiling its predecessor took two months to reach.

That said, single-day ELO readings carry higher variance. The 1505 figure will settle further over coming weeks as vote volume increases. Gemini 3.1 Pro’s position is also evolving — the model entered earlier and has more stabilised ratings.

Pricing Gap

Both models at the top of the general leaderboard carry very different price tags. Claude Opus 4.7 runs at $5 per million input tokens and $25 per million output tokens. Gemini 3.1 Pro Preview is priced at $2 input / $12 output. At equal ELO, Gemini retains a 60% input cost advantage over Claude.

That differential is consistent with the intelligence-to-cost analysis published earlier this month: Gemini 3.1 Pro was described as matching GPT-5.4 on intelligence at 20% lower cost. Against Claude Opus 4.7, the cost gap is larger still.

Key Numbers

  • Claude Opus 4.7 Thinking Arena ELO: 1505 (Coding: 1565)
  • Claude Opus 4.7 standard Arena ELO: 1503 (Coding: 1554)
  • Gemini 3.1 Pro Preview Arena ELO: 1505 (Coding: 1531)
  • GPT-5.4 High Arena ELO: 1495 (Coding: 1538)
  • Claude Opus 4.7 pricing: $5/$25 per million tokens
  • Gemini 3.1 Pro pricing: $2/$12 per million tokens