BenchLeader
Nous ResearchOpen weightsReasoning modelCompare with another model

Hermes 4 70B

Reasoning effort

As of 16 Nov 2026, Hermes 4 70B by Nous Research is not yet ranked on the BenchLeader Index, which needs at least five independent results across two categories. Released 27 Aug 2025.

Key facts (as of 16 Nov 2026)

  • Hermes 4 70B was released on 27 Aug 2025 (Nous Research).

In detail. Hermes 4 70B is a Nous Research open-weights reasoning model, released 27 Aug 2025. It is not yet ranked: the index needs at least five independent results across two categories. It has been measured at 2 reasoning-effort settings; this summary describes the best-scoring one, and the tabs above switch between them. Last measured 27 Aug 2025.stale: no new result in six months

Blended price
–
Output speed
–
First answer
–
Context
128k
via Artificial Analysis
Cost per task
–
Tokens per task
–
Released
27 Aug 2025
How it scores by categoryDashed line = average model (50). One step of 15 = one standard deviation.
  1. Agents & tools34
  2. Instruction following33
  3. Long context25

Get an email when Hermes 4 70B gets its first benchmark results

And when Nous Research ships a new model, plus the weekly review on Mondays.

Choose providers and emails

Reasoning-effort configurations

The same model behaves differently depending on how much it is allowed to think. Each row is one setting, scored only on the benchmarks that were run at that setting. “Not stated” collects results from publishers that did not say which setting they used; for a reasoning model that is usually its thinking mode, but we do not assume it. Pick a setting here or at the top of the page to see its price, speed and category scores.

EffortIndexRankSpeedFirst answerChat reply costCategories
no reasoningbest–––––Agents & tools 34 · Instruction following 33 · Long context 25
thinking–––––Agents & tools 38 · Instruction following 35 · Long context 28

Benchmark results

See all LLM benchmarks. One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.

Metadata

Benchmarkno reasoningthinkingSource
AA Openness Indexnot in index47.2#9247.2#92Artificial Analysis

Reasoning

Benchmarkno reasoningthinkingSource
GPQA Diamond (AA)not in index49.1%#44069.9%#305Artificial Analysis
Humanity's Last Exam (AA)not in index3.6%#6098.8%#363Artificial Analysis
CritPtnot in index0.0%#3370.0%#337Artificial Analysis

Coding

Benchmarkno reasoningthinkingSource
LiveCodeBench (AA)not in index26.9%#22065.3%#90Artificial Analysis

Agents & tools

Benchmarkno reasoningthinkingSource
Terminal-Bench Hard0.0%#3674.5%#289Artificial Analysis
τ²-Bench Telecom (AA)not in index21.6%#32422.5%#316Artificial Analysis

Maths

Benchmarkno reasoningthinkingSource
AIME 2025 (AA)not in index11.3%#21868.7%#95Artificial Analysis

Instruction following

Benchmarkno reasoningthinkingSource
IFBench29.0%#37531.3%#361Artificial Analysis

Long context

Benchmarkno reasoningthinkingSource
AA-LCR v1.14.7%#5139.7%#493Artificial Analysis

Composite

Benchmarkno reasoningthinkingSource
AA Intelligence Index v4.3.2not in index6.7#5267.9#459Artificial Analysis
AA Progress Index v1not in index156.4#546199.7#418Artificial Analysis

More from Nous Research

New this month

See also

Questions people ask

Why does Hermes 4 70B have no BenchLeader Index score?
The index needs at least five independent results across at least two categories, and as of 16 Nov 2026 Hermes 4 70B does not have them yet. Launch-post scores are never used, so it is ranked once independent publishers have measured it.
When was Hermes 4 70B released?
Nous Research released Hermes 4 70B on 27 Aug 2025. BenchLeader has results for it at 2 reasoning-effort settings.

Data as of 16 Nov 2026. Compare these configurations.

Cite as: BenchLeader, “Hermes 4 70B: benchmarks, pricing, speed and rank”, data as of 16 Nov 2026, https://www.benchleader.com/models/hermes-4-llama-3-1-70b.