Benchmark Almanac

Vals Web Search Index

Claude Opus 5.5 57.5% leads since September 2026
Not in model cards

Agents work Finance Agent v2 and Legal Research Bench tasks with one web search tool as their only outside source. It compares native provider search with Exa and Keenable.

Domain tools such as SEC EDGAR and CourtListener are removed. The board had nine model and tool pairs on October 21st, 2026.

Made by
Vals AI
Released
July 2026
Kind
Deep research
Level
Expert
Scored
Graded by a model
Answers
Agent task
Tasks
Kept private
Strongest model tested
Claude Opus 5.5 #4 on the capability index
Last model added
September 2026
Scores refreshed
November 12th, 2026
5
Models scored on its own leaderboard

Claude Opus 5.5 leads with 57.5% among models released by September 2026

Best score per model from the benchmark's own leaderboard, read on November 12th, 2026. Where a model ran with several settings or harnesses, we keep its best. Every run is in the next chart.

The newest model with a score here came out in September 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 57.5 Claude Opus 5.5
  2. 49.1 GPT-5.6 Sol
  3. 48.5 Claude Fable 5
  4. 41.4 Gemini 3.5 Flash
  5. 38.8 Grok 4.5
  1. 1 Claude Opus 5.5 57.5%
  2. 2 GPT-5.6 Sol 49.1%
  3. 3 Claude Fable 5 48.5%
  4. 4 Gemini 3.5 Flash 41.4%
  5. 5 Grok 4.5 38.8%

Source: Official Vals Web Search Index leaderboard

#ModelScoreSettingDateSource
1 Claude Opus 5.5 57.5% Parallel Fast Sep 2026 Official leaderboard
2 GPT-5.6 Sol 49.1% Keenable Jul 2026 Official leaderboard
3 Claude Fable 5 48.5% Exa Jun 2026 Official leaderboard
4 Gemini 3.5 Flash 41.4% Exa May 2026 Official leaderboard
5 Grok 4.5 38.8% Exa Jul 2026 Official leaderboard

Search score vs. cost

The line joins the runs that no other run beats on both measures. We redrew this chart from the numbers on the official leaderboard.

40% 50% 60% $0$5$10$15$20 Cost per task ($), lower is better Score Gemini 3.5 Flash: Score 41.4%, cost per task $1.18 Claude Fable 5: Score 48.5%, cost per task $3.48 Claude Fable 5: Score 46.9%, cost per task $4.37 Grok 4.5: Score 35.5%, cost per task $0.99 GPT-5.6 Sol: Score 49.1%, cost per task $8.84 GPT-5.6 Sol: Score 45.2%, cost per task $19 Claude Opus 5.5: Score 53.8%, cost per task $3.48 Gemini 3.5 Flash: Score 37%, cost per task $0.5 Grok 4.5: Score 38.8%, cost per task $0.58 GPT-5.6 Sol: Score 43.6%, cost per task $0.95 Claude Opus 5.5: Score 57.5%, cost per task $2.27 Claude Opus 5.5: Score 54%, cost per task $2.21 Claude Opus 5.5Claude Opus 5.5GPT-5.6 SolGrok 4.5Gemini 3.5 Flash 40% 50% 60% $0$5$10$15$20 Cost per task ($), lower is better Score Gemini 3.5 Flash: Score 41.4%, cost per task $1.18 Claude Fable 5: Score 48.5%, cost per task $3.48 Claude Fable 5: Score 46.9%, cost per task $4.37 Grok 4.5: Score 35.5%, cost per task $0.99 GPT-5.6 Sol: Score 49.1%, cost per task $8.84 GPT-5.6 Sol: Score 45.2%, cost per task $19 Claude Opus 5.5: Score 53.8%, cost per task $3.48 Gemini 3.5 Flash: Score 37%, cost per task $0.5 Grok 4.5: Score 38.8%, cost per task $0.58 GPT-5.6 Sol: Score 43.6%, cost per task $0.95 Claude Opus 5.5: Score 57.5%, cost per task $2.27 Claude Opus 5.5: Score 54%, cost per task $2.21 Claude Opus 5.5Claude Opus 5.5GPT-5.6 SolGrok 4.5Gemini 3.5 Flash
ModelSettingScoreCost per task
Claude Opus 5.5 Parallel Fast 57.5% $2.27
Claude Opus 5.5 Parallel Advanced 54.1% $2.21
Claude Opus 5.5 Native 53.8% $3.48
GPT-5.6 Sol Keenable 49.1% $8.84
Claude Fable 5 Exa 48.5% $3.48
Claude Fable 5 Native 46.9% $4.37
GPT-5.6 Sol Exa 45.2% $18.97
GPT-5.6 Sol Native 43.6% $0.95
Gemini 3.5 Flash Exa 41.4% $1.18
Grok 4.5 Exa 38.8% $0.58
Gemini 3.5 Flash Native 37% $0.5
Grok 4.5 Native 35.5% $0.99

The best score, 57.5%, costs $2.27 a task

Runs on the cost frontier score higher than any run that's cheaper. The costs are the leaderboard's own estimates, so compare them only within this benchmark.

Google DeepMind xAI OpenAI Anthropic Cost frontier One model at each effort level
0% 20% 40% 60% $0.10 $1.00 $10 $100 Cost per task, log scale Claude Opus 5.5 Claude Opus 5.5 GPT-5.6 Sol Grok 4.5 0% 20% 40% 60% $0.10 $1.00 $10 $100 Cost per task, log scale Claude Opus 5.5 Claude Opus 5.5

Source: Official Vals Web Search Index leaderboard

ModelEffortHarnessScoreCost per taskOutput tokens
Claude Opus 5.5 Parallel Fast 57.5% $2.27
Claude Opus 5.5 Parallel Advanced 54% $2.21
Claude Opus 5.5 Native 53.8% $3.48
GPT-5.6 Sol Keenable 49.1% $8.84
Claude Fable 5 Exa 48.5% $3.48
Claude Fable 5 Native 46.9% $4.37
GPT-5.6 Sol Exa 45.2% $19
GPT-5.6 Sol Native 43.6% $0.95
Gemini 3.5 Flash Exa 41.4% $1.18
Grok 4.5 Exa 38.8% $0.58
Gemini 3.5 Flash Native 37% $0.50
Grok 4.5 Native 35.5% $0.99

3 models have held number one on the official leaderboard since its release in July 2026

Lab claims count from the day the card came out, independent runs from the model’s release.

Anthropic OpenAI
Own leaderboard Not tested since Sep 2026
Claude Opus 5.5 57.5%

Source: Official Vals Web Search Index leaderboard

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

On the benchmark’s own leaderboard Best so far
0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Claude Opus 5.5 GPT-5.6 Sol Gemini 3.5 Flash 0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Claude Opus 5.5 GPT-5.6 Sol

Source: Official Vals Web Search Index leaderboard

No lab model card reports it yet

We read the headline results of 522 launch posts from 52 labs, and none of them list this benchmark.

More in Tool use

All 144 in Tool use
  • AutomationBench In use Agent completes end-to-end business workflows inside a simulated company exposing dozens of REST APIs. Graded on resulting state. Gemini 4 Carbon 53.9%
  • MCP Atlas In use Multi-step workflows over real MCP servers. Scores tool-use competency on realistic tasks. Gemini 4 Argon 90.7%
  • BFCL V4 In use Models call functions and tools, including multi-turn agentic tasks with web search and memory. Checked against expected calls and outcomes. Qwen3.7-Max 75%
  • τ³-bench Banking In use Agents act as banking customer-service reps over many turns with a simulated user, following policy. Pass^1 on the final database state. Qwen 3.8 Max 55.1%
  • Toolathlon-Verified In use 108 long-horizon tasks across 32 apps through about 600 MCP tools. Execution-based checkers inspect final environment state. GLM-5.3-Flash 78.4%
  • DeepSearchQA In use Questions that need multi-step search and synthesis. Final answers graded against references. Kimi K3 95%