OpenHermes-2.5-Mistral-7B

7.24B parameters, text in, text out, 32K tokens of context. 1 provider serves it.

Verified 2026-10-25 Rank #397 in LLMs · score 79.7 method v1 Sources 3 Downloads 30d 4.6k · #551 of 619 tracked

Can I run it?4.3GB at INT4Estimatefits a 16 GB cardmethod

Can I use it?Apache 2.0Permissive; patent grantlicence file2026-10-24

Do people use it?4.6kdownloads, 30 days#551 of 619 we track · below the median+2 likes, downloads −7% this week921 likesHugging Face2026-10-24

Worth reading

Picked by rule from 2 tracked pieces: the lab's own post, the most shared, a field report, the latest.

Latest · 3 years ago

Open Hermes 2.5 and Nous-Hermes 2 dataset now public

twitter.com 2024-01-31 3 HN pts about the OpenHermes 2.5 family family

2 best of 2 tracked

The 1 pick above, plus these 1. Top is the radar's score: discussion, source authority and the lab's own posts, fading with age.

DiscussionOpenHermes-2.5-Neuralchat-Slerptwitter.com · 2023-12-09 · 2 HN pts

Where to run it

Live hosted endpoints, cheapest first, USD per million tokens. Ties break on output price, then provider name.

#	Cheapest live row	In /1M	Out /1M	Context	Quant	Seen
1	Featherless	—	—	—	—	Oct 25

All 1 provider →

Where it's used

Hugging Face does not say who downloads a model or with what. These are the proxies: API providers listing it, quantized builds, fine-tunes/adapters/merges, and Spaces that load it.

Served by 1 API provider; heavily run locally (53 quantized builds); a big base for builders (499 fine-tunes, adapters and merges).

As a service1Featherless
On own hardware53quantized builds
Built on49983 fine-tunes · 366 adapters · 50 merges
In apps88Hugging Face Spaces

Top Spaces running it: aimevzulari/Prompt_Uzmani ♥ 46 · artificialguybr/OPENHERMES-V2.5-DEMO ♥ 17 · EmbeddedLLM/chat-template-generation ♥ 12 · wenkai/FAPM_demo ♥ 11

Will it fit?

Weights only, an estimate: leave room for the KV cache. Method

Fits a 16 GB card at INT4

About 4.3 GB at INT4, 8.7 GB at FP8, 17 GB at BF16: weights only, so leave room for the KV cache.

Estimate params × bytes per param × 1.2 · how we estimate

The story so far

Weights, every dated article, and repo commits on one rail.

2 dated articles and 17 repo commits so far

Teal is the weights (2023-10-29), amber triangles are 2 dated articles (each links to it), grey ticks are repo commits.

What people built on it

552 derivatives on the Hub. Which build to run, by runtime: the most downloaded in each class.

552 derivatives on the Hub, most of them adapters

From the Hugging Face model tree; quantized builds in teal, fine-tunes, merges and adapters in greys.

GGUF TheBloke/OpenHermes-2.5-Mistral-7B-GGUF for llama.cpp, Ollama, LM Studio 499.2k downloads AWQ TheBloke/OpenHermes-2.5-Mistral-7B-AWQ for vLLM, SGLang 457.8k downloads

All variants →

Reference

Everything else, folded: the full tables, every source.

Benchmarks

41 rows

Benchmarks 41 rows

measured third-party leaderboard rowauthor-reported number from the model card or eval files (41)No confidence intervals unless the source gave them; settings differ between rows, so compare within a benchmark, not across. Date = the benchmark's own date when it gives one, else when we read it.

Benchmark	Value	Setting / notes	Reported by	Date
author-reported AGI-Eval	43.07% percent	—	model author	2026-10-07
author-reported ARC Challenge · Chat and follow instructions	0.56	GPT4All Version 0	model author	2026-10-07
author-reported Average Total	52.38	—	model author	2026-10-07
author-reported BigBench Reasoning Test	40.96% percent	—	model author	2026-10-07
author-reported Commonsense QA sets · Chat and follow instructions	0.83	GPT4All Version 0	model author	2026-10-07
author-reported Commonsense QA sets · Chat and follow instructions	0.87	GPT4All Version 1	model author	2026-10-07
author-reported Commonsense QA sets · Chat and follow instructions	0.35	GPT4All Version 0	model author	2026-10-07
author-reported Commonsense QA sets · Chat and follow instructions	0.81	GPT4All Version 0	model author	2026-10-07
author-reported GPT4All Benchmark Set	73.12	—	model author	2026-10-07
author-reported HellaSwag · Chat and follow instructions	0.63	GPT4All Version 0	model author	2026-10-07
author-reported HumanEval · Write code with agents	50.7% pass@1	—	model author	2026-10-07
author-reported MATH-500 · Reason through hard problems	0.38	AGI-Eval Version 0	model author	2026-10-07
author-reported Total Score	210.19	—	model author	2026-10-07
author-reported TruthfulQA · Chat and follow instructions	53.04	—	model author	2026-10-07
author-reported TruthfulQA · Chat and follow instructions	0.36	TruthfulQA Version 1	model author	2026-10-07
author-reported Winogrande · Chat and follow instructions	0.74	GPT4All Version 0	model author	2026-10-07
author-reported agieval_aqua_rat	0.23	AGI-Eval Version 0	model author	2026-10-07
author-reported agieval_logiqa_en	0.39	AGI-Eval Version 0	model author	2026-10-07
author-reported agieval_lsat_ar	0.25	AGI-Eval Version 0	model author	2026-10-07
author-reported agieval_lsat_lr	0.51	AGI-Eval Version 0	model author	2026-10-07
author-reported agieval_lsat_rc	0.59	AGI-Eval Version 0	model author	2026-10-07
author-reported agieval_sat_en	0.75	AGI-Eval Version 0	model author	2026-10-07
author-reported agieval_sat_en_without_passage	0.44	AGI-Eval Version 0	model author	2026-10-07
author-reported bigbench_causal_judgement	0.53	BigBench Version 0	model author	2026-10-07
author-reported bigbench_date_understanding	0.67	BigBench Version 0	model author	2026-10-07

Show all 41 rows (16 more)

Benchmark	Value	Setting / notes	Reported by	Date
author-reported bigbench_disambiguation_qa	0.34	BigBench Version 0	model author	2026-10-07
author-reported bigbench_geometric_shapes	0.21	BigBench Version 0	model author	2026-10-07
author-reported bigbench_logical_deduction_five_objects	0.29	BigBench Version 0	model author	2026-10-07
author-reported bigbench_logical_deduction_seven_objects	0.21	BigBench Version 0	model author	2026-10-07
author-reported bigbench_logical_deduction_three_objects	0.48	BigBench Version 0	model author	2026-10-07
author-reported bigbench_movie_recommendation	0.36	BigBench Version 0	model author	2026-10-07
author-reported bigbench_navigate	0.5	BigBench Version 0	model author	2026-10-07
author-reported bigbench_reasoning_about_colored_objects	0.66	BigBench Version 0	model author	2026-10-07
author-reported bigbench_ruin_names	0.42	BigBench Version 0	model author	2026-10-07
author-reported bigbench_salient_translation_error_detection	0.23	BigBench Version 0	model author	2026-10-07
author-reported bigbench_snarks	0.68	BigBench Version 0	model author	2026-10-07
author-reported bigbench_sports_understanding	0.65	BigBench Version 0	model author	2026-10-07
author-reported bigbench_temporal_sequences	0.28	BigBench Version 0	model author	2026-10-07
author-reported bigbench_tracking_shuffled_objects_five_objects	0.21	BigBench Version 0	model author	2026-10-07
author-reported bigbench_tracking_shuffled_objects_seven_objects	0.17	BigBench Version 0	model author	2026-10-07
author-reported bigbench_tracking_shuffled_objects_three_objects	0.48	BigBench Version 0	model author	2026-10-07

What the card says

1 cited sentence

Summary generated, cited

It was trained on 1,000,000 entries of primarily GPT-4 generated data.^[1]

“trained on 1,000,000 entries of primarily GPT-4 generated…”

Generated 2026-10-07 15:09 UTC by parasail-gemma-4-26b-a4b-it; each numeral points at the fact it rests on. Not a verdict.

Use-case evidence

4 use cases

What it is good at 4 use cases

Labels are evidence, not verdicts: measured a third-party leaderboard row for this model; author-reported a number the model author published in the card or eval files; structural a deterministic signal from the name, config or chat template, or a decision-model estimate. Each chip links to its source. All use cases.

Chat and follow instructionsauthor-reportedAuthor reports TruthfulQA 53.04 Author reports arc_challenge 0.5623 Author reports boolq 0.8657 +5 more
Write code with agentsauthor-reportedAuthor reports HumanEval 50.7%
Reason through hard problemsauthor-reportedAuthor reports agieval_sat_math 0.3773
Run on a laptop or phonestructuraldecision-model estimate p=0.99 fits: single-gpu

Facts

20 fields

Listed in LLMs

Facts

OpenHermes-2.5-Mistral-7B has 7.24B parameters. (safetensors metadata, observed 2026-10-24)
Context length is 32,768 tokens (32K). (config.json max_position_embeddings, observed 2026-10-24)
License: Apache 2.0 (friction: none). Permissive; patent grant. (license file, observed 2026-10-24)
Created 2023-10-29, last modified 2024-02-19. (HF API, observed 2026-10-24)

Pipeline tag	text-generation
Library	transformers
Architectures	MistralForCausalLM
Model type	mistral
Dtype	bfloat16
Params by dtype	BF16: 7,241,748,480
Size band	M
MoE	no
Context length	32,768 (config.json max_position_embeddings)
Languages	en
License	Apache 2.0 (apache-2.0)
Hub snapshot	2026-10-25

Servability

vLLM · SGLang

Servability

vLLM: supported · MistralForCausalLM · 7050ef881ff0835f6890a47db1859a24c65481d6SGLang: supported · MistralForCausalLM · 773cb938fe9484a8f7e7e46fb1b1fe96412c59ed

vLLM registry (7050ef881ff0835f6890a47db1859a24c65481d6) · SGLang registry (773cb938fe9484a8f7e7e46fb1b1fe96412c59ed) · Matched by architecture class name against each project's registry on main; 'supported' means the architecture is registered, not that this checkpoint was tested.

Hardware fit

3 estimates

Hardware fit ESTIMATE

Weights-only memory for OpenHermes-2.5-Mistral-7B: params × bytes/param × 1.2 overhead. KV cache and activations are extra. How this is computed.

BF16 (2 bytes/param)	17 GB · Single GPU (<24 GB)
FP8 (1 byte/param)	8.7 GB · Single GPU (<24 GB)
INT4 (0.5 bytes/param)	4.3 GB · Single GPU (<24 GB)

Weights only (params × bytes/param × 1.2 overhead); KV cache and activations extra.

All variants

40 variants

Lineage

Fine-tuned from mistralai/Mistral-7B-v0.1 Family: teknium/OpenHermes-2.5 552 derivatives on the Hub

Base model: mistralai/Mistral-7B-v0.1

quantized: 53finetune: 83merge: 50adapter: 366

Top variants (40 shown, ranked by downloads)

Variant	Relation	Quant	Params	Band	License	Author	Likes	Downloads
TheBloke/OpenHermes-2.5-Mistral-7B-GGUF	quantized	GGUF	—	—	apache-2.0	TheBloke	283	499.2k
TheBloke/OpenHermes-2.5-Mistral-7B-AWQ	quantized	AWQ	7.24B	M	apache-2.0	TheBloke	22	457.8k
TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ	quantized	GPTQ	7.24B	M	apache-2.0	TheBloke	29	54.4k
mlabonne/NeuralHermes-2.5-Mistral-7B	finetune	—	7.24B	M	apache-2.0	mlabonne	151	36.1k
sayhan/OpenHermes-2.5-Strix-Philosophy-Mistral-7B-LoRA	quantized	GGUF	7.24B	M	apache-2.0	sayhan	9	28.4k
EmbeddedLLM/Mistral-7B-Merge-14-v0.3	merge	—	7.24B	M	apache-2.0	EmbeddedLLM	8	19.2k
openaccess-ai-collective/DPOpenHermes-7B-v2	finetune	—	7.24B	M	apache-2.0	openaccess-ai-collective	31	17.1k
Weyaxi/OpenHermes-2.5-neural-chat-v3-3-Slerp	merge	—	7.24B	M	apache-2.0	Weyaxi	41	15.4k
argilla/CapybaraHermes-2.5-Mistral-7B	finetune	—	7.24B	M	apache-2.0	argilla	73	13.6k
EmbeddedLLM/Mistral-7B-Merge-14-v0.3-ft-step-15936	merge	—	7.24B	M	apache-2.0	EmbeddedLLM	0	17k
EmbeddedLLM/Mistral-7B-Merge-14-v0.3-ft-step-9984	merge	—	7.24B	M	apache-2.0	EmbeddedLLM	0	17k
beowolx/MistralHermes-CodePro-7B-v1	finetune	—	7.24B	M	mit	beowolx	14	15.8k
second-state/OpenHermes-2.5-Mistral-7B-GGUF	quantized	GGUF	—	—	apache-2.0	second-state	4	16.1k
EmbeddedLLM/Mistral-7B-Merge-02-v0	merge	—	7.24B	M	apache-2.0	EmbeddedLLM	0	15.9k
nlpguy/Hermes-low-tune-2	merge	—	7.24B	M	apache-2.0	nlpguy	2	13.6k
gqd/mistral-merge-7b	merge	—	7.24B	M	unlicense	gqd	1	13.5k
RedHatAI/OpenHermes-2.5-Mistral-7B-pruned50	finetune	—	—	—	—	RedHatAI	1	13.1k
Gille/StrangeMerges_18-7B-dare_ties	merge	—	7.24B	M	apache-2.0	Gille	0	13.2k
RedHatAI/OpenHermes-2.5-Mistral-7B-marlin	quantized	—	1.19B	S	—	RedHatAI	2	12.5k
nlpguy/Hermes-low-tune-3.1	merge	—	7.24B	M	—	nlpguy	0	12.3k
rxavier/Taurus-7B-1.0	finetune	—	7.24B	M	mit	rxavier	1	12.1k
openaccess-ai-collective/openhermes-2_5-dpo-no-robots	finetune	—	7.24B	M	apache-2.0	openaccess-ai-collective	11	11.5k
openaccess-ai-collective/DPOpenHermes-7B	finetune	—	7.24B	M	apache-2.0	openaccess-ai-collective	16	11k
RedHatAI/OpenHermes-2.5-Mistral-7B-pruned2.4	finetune	—	—	—	—	RedHatAI	0	11.5k
saishf/West-Hermes-7B	merge	—	7.24B	M	apache-2.0	saishf	4	11.3k
joey00072/ToxicHermes-2.5-Mistral-7B	finetune	—	7.24B	M	apache-2.0	joey00072	22	10.3k
mayacinka/chatty-djinn-14B	merge	—	13.6B	L	apache-2.0	mayacinka	0	10.9k
giraffe176/Open_Maid_Samantha_Hermes_Orca_dare_tiesv0.1	merge	—	7.24B	M	cc-by-4.0	giraffe176	0	10.9k
giraffe176/WestLake_Noromaid_OpenHermes_neural-chatv0.1	merge	—	7.24B	M	cc-by-nc-4.0	giraffe176	2	10.5k
hydra-project/OpenHyperion-2.5-Mistral-7B	merge	—	7.24B	M	apache-2.0	hydra-project	2	10.1k
giraffe176/Open_Neural_Monarch_Maidv0.1	merge	—	7.24B	M	cc-by-nc-4.0	giraffe176	0	10.1k
MaziyarPanahi/OpenHermes-2.5-Mistral-7B-GGUF	quantized	GGUF	—	—	apache-2.0	MaziyarPanahi	2	10k
openaccess-ai-collective/dpopenhermes-alpha-v0	finetune	—	7.24B	M	apache-2.0	openaccess-ai-collective	4	9.6k
mlabonne/NeuralHermes-2.5-Mistral-7B-laser	finetune	—	7.24B	M	apache-2.0	mlabonne	16	8.8k
bartowski/OpenHercules-2.5-Mistral-7B-GGUF	merge	GGUF	—	—	—	bartowski	0	9.5k
charlesdedampierre/TopicNeuralHermes-2.5-Mistral-7B	finetune	—	7.24B	M	apache-2.0	charlesdedampierre	6	8.5k
decruz07/kellemar-DPO-7B-v1.01	finetune	—	7.24B	M	apache-2.0	decruz07	3	8.1k
decruz07/kellemar-DPO-7B-d	finetune	—	7.24B	M	apache-2.0	decruz07	0	8k
hydra-project/OpenHercules-2.5-Mistral-7B	merge	—	7.24B	M	apache-2.0	hydra-project	7	6.4k
QuantFactory/OpenHermes-2.5-Mistral-7B-GGUF	quantized	GGUF	—	—	apache-2.0	QuantFactory	0	6.7k

Repository changelog

17 commits

Repository changelog 17 commits

Date	Commit	SHA
2024-02-19	Update README.md	24c0bea
2023-12-03	Adding `safetensors` variant of this model (#10)	91ed666
2023-11-29	Update README.md	3b083f8
2023-11-24	Update README.md	36288de
2023-11-04	Update transformers_inference.py	2a54cad
2023-11-03	Update transformers_inference.py	9ae7036
2023-11-03	add chat template	93f2ad1
2023-11-03	Create transformers_inference.py	359079f
2023-11-03	Update README.md	8820e1a
2023-11-02	Update README.md	408d5b9
2023-11-02	Update README.md	56cbbc6
2023-11-02	Update README.md	8eacbab
2023-11-02	Update README.md	d4a2212
2023-11-01	Update README.md	d583c44
2023-11-01	Create README.md	2beae5b
2023-10-29	Upload 9 files	7346b49
2023-10-29	initial commit	1cd80a3

All providers

1 provider · 1 endpoint

Where to run 1 providers · 1 endpoints

Hosted endpoints observed for OpenHermes-2.5-Mistral-7B, sorted by input price (USD per million tokens). Each row names its observer and date; prices change daily, so check the source before quoting.

Provider	Model ref	Context	Quant	$/M in	$/M out	tok/s	TTFT	Tools	Status	Observed
Featherless hf	teknium/OpenHermes-2.5-Mistral-7B	—	—	—	—	—	—	—	live	2026-10-25

Articles and coverage

2 pieces

Ranked by source signals (HN, Lobsters, authority domains, official posts) and recency.

Community2

Open Hermes 2.5 and Nous-Hermes 2 dataset now public

twitter.com2024-01-31about the OpenHermes 2.5 familyHN 3 pts · 0 comments

OpenHermes-2.5-Neuralchat-Slerp

twitter.com2023-12-09about the OpenHermes 2.5 familyHN 2 pts · 0 comments

Similar models

3 models

Similar models

Llama-3.1-8B-Instruct · 8.03B
same size band (M), same task (text-generation), more providers
Qwen2.5-7B-Instruct · 7.62B
same size band (M), same task (text-generation), more providers, permissive license
Mistral-7B-Instruct-v0.2 · 7.24B
same size band (M), same task (text-generation), permissive license

Sources

3 sources

Sources

Hub snapshot: 2026-10-25 · HF API: 2026-10-24 21:38 UTC · Providers: 2026-10-24 21:44 UTC · Articles: 2026-10-24 21:46 UTC · Summary: 2026-10-07 15:09 UTC · Last verified 2026-10-25

Follow OpenHermes-2.5-Mistral-7B

Watch this model

/feeds/models/teknium--OpenHermes-2.5-Mistral-7B.xml(RSS 2.0) · every change to OpenHermes-2.5-Mistral-7B
/feeds/families/teknium--OpenHermes-2.5.xml(RSS 2.0) · every change across the OpenHermes-2.5 family

Email alerts are not open yet; the feeds above are the same data. See feeds for the item schema.