Benchmark Almanac

Speko Speech-to-Speech Benchmark (10-turn call)

grok-voice-fast 1111% tops its leaderboard Gemini 3.1 Flash Live 1108% leads the models we track
Not in model cards
Version 10-turn

Voice agents answer a weather request with a tool, handle an interruption, and recall an earlier detail. A language model grades task success.

Speko published task success and response time for nine ranked models across 15 sessions per model. It excluded a tenth model that spoke its tool calls.

Made by
Speko
Released
July 2026
Kind
Customer service
Level
Everyday
Scored
Graded by a model
Answers
Agent task
Measures
Capability
Language
English
In and out
Audio, Text → Audio, Actions
Interaction
Agentic
Environment
APIs and tools
Items from
Written by experts
Run by
A host
Last model added
July 2026
Scores refreshed
November 20th, 2026
9
Entries on its own leaderboard, 6 models we track

grok-voice-fast tops the leaderboard with 1111% Gemini 3.1 Flash Live leads with 1108% among models released by July 2026

Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. The best model we track is Gemini 3.1 Flash Live, at 1108%. This view shows only models we track, each at its best.

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

  1. 1 grok-voice-fast 1111%
  2. 2 Gemini 3.1 Flash Live 1108%
  3. 3 GPT-Realtime-2.1 1104%
  4. 4 GPT-Realtime-2 1103%
  5. 5 GPT-Realtime-2.1 Mini 1011%
  6. 6 grok-voice-think-fast 1007%
  7. 7 Inworld Realtime 973%
  8. 8 GPT-Realtime mini Dec 2025 614%
  9. 9 gpt-realtime 494%

Source: Official Speko Speech-to-Speech Benchmark (10-turn call) leaderboard

#EntryScoreSettingReleasedRun onSource
1 grok-voice-fast 1111% Jul 2026 Official leaderboard
2 Gemini 3.1 Flash Live 1108% Mar 2026 Jul 2026 Official leaderboard
3 GPT-Realtime-2.1 1104% Jul 2026 Jul 2026 Official leaderboard
4 GPT-Realtime-2 1103% May 2026 Jul 2026 Official leaderboard
5 GPT-Realtime-2.1 Mini 1011% Jul 2026 Jul 2026 Official leaderboard
6 grok-voice-think-fast 1007% Jul 2026 Official leaderboard
7 Inworld Realtime 973% Jul 2026 Official leaderboard
8 GPT-Realtime mini Dec 2025 614% Dec 2025 Jul 2026 Official leaderboard
9 gpt-realtime 494% Aug 2025 Jul 2026 Official leaderboard
#ModelScoreSettingReleasedRun onSource
1 Gemini 3.1 Flash Live 1108% Mar 2026 Jul 2026 Official leaderboard
2 GPT-Realtime-2.1 1104% Jul 2026 Jul 2026 Official leaderboard
3 GPT-Realtime-2 1103% May 2026 Jul 2026 Official leaderboard
4 GPT-Realtime-2.1 Mini 1011% Jul 2026 Jul 2026 Official leaderboard
5 GPT-Realtime mini Dec 2025 614% Dec 2025 Jul 2026 Official leaderboard
6 gpt-realtime 494% Aug 2025 Jul 2026 Official leaderboard

Number one over time

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

Google DeepMind
Own leaderboard Passed 85% in Jul 2026, by Gemini 3.1 Flash Live

Source: Official Speko Speech-to-Speech Benchmark (10-turn call) leaderboard

Gpt-realtime on the official leaderboard passed 85% in August 2025

Models released before it came out already scored that high. All the scores we have, one dot each, so a model tested at several settings appears more than once.

Plot with other benchmarks
On the benchmark’s own leaderboard Best so far
0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Gemini 3.1 Flash Live GPT-Realtime mini gpt-realtime 0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Gemini 3.1 Flash Live

Source: Official Speko Speech-to-Speech Benchmark (10-turn call) leaderboard

No lab model card reports it yet

We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.

Versions

More in Tool use

All 265 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 Neon 56.3%
  • 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%

Data audit

8 of 9 checks pass, 1 warning. Look first at the Task access check. We don’t know whether the tasks are public. Every benchmark

Catalog record 1 warning 1 pass
  • Warning Task access We don’t know whether the tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 3 pass 2 notes
  • Pass Independent runs Scores from the official leaderboard (9 entries).
  • Pass Models scored 6 models we track, plus 3 other entries on its leaderboard.
  • Pass Newest score The newest is from July 2026, for GPT-Realtime-2.1.
  • Note Frontier models None of its models are on the capabilities index.
  • Note Headroom Saturated. The best is 1108%, past 85% since August 2025.
Leaderboard import 3 pass
  • Pass Importer We read the board on speko.com: 6 rows for models we track and 3 other entries at the last read.
  • Pass Last read November 20th, 2026.
  • Pass Read worked The last read worked.
Consistency 1 pass 1 note
  • Note Lab claims No model has both a lab number and an independent run.
  • Pass Known issues No published audit found problems.

How we check