2022

Constitutional AI (CAI) introduced in paper: Constitutional AI: Harmlessness from AI Feedback (Bai et al., 2022).

ba1eb83c14795107 · 2 sources · 100% confidence

InstructGPT methodology introduced in paper: Training language models to follow instructions with human feedback (Ouyang et al., 2022).

5da8f8dffc038b8e · 2 sources · 100% confidence

FlashAttention introduced in paper: FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness (Dao et al., 2022).

e120182d1e01ea2b · 2 sources · 100% confidence

ChatGPT released on: 2022-11-30.

8d653880c519a8ef · 2 sources · 100% confidence

Whisper released on: 2022-09-21.

a3ebbaed14bd83d0 · 2 sources · 100% confidence

Chinchilla scaling laws introduced in paper: Training Compute-Optimal Large Language Models (Hoffmann et al., 2022).

8befcae6bce01a95 · 2 sources · 100% confidence

Chain-of-Thought prompting introduced in paper: Chain-of-Thought Prompting Elicits Reasoning in Large Language Models (Wei et al., 2022).

3af924da138ff84c · 2 sources · 100% confidence

Stable Diffusion 1.0 released on: 2022-08-22.

f6f333228f224df2 · 2 sources · 100% confidence

ReAct (Reasoning + Acting) introduced in paper: ReAct: Synergizing Reasoning and Acting in Language Models (Yao et al., 2022).

fceea64fa7d04d3a · 2 sources · 100% confidence

DALL·E 2 released on: 2022-04-06.

0b0e64476bd25bd6 · 2 sources · 100% confidence

Stable Diffusion 1.x released on: 2022-08-22.

79a7e980e59680bc · 2 sources · 100% confidence

Together AI founded in: 2022.

4d05fa0726841c32 · 2 sources · 90% confidence

PaLM introduced in paper: PaLM: Scaling Language Modeling with Pathways (Chowdhery et al., 2022).

d58d505fd9d705fe · 2 sources · 100% confidence

Imagen introduced in paper: Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding (Saharia et al., 2022).

30fdfa95f8684ca5 · 2 sources · 100% confidence

Flamingo introduced in: Alayrac et al. 2022 — DeepMind few-shot vision-language model.

72ea74efc723bd06 · 2 sources · 100% confidence

GPTQ introduced in: Frantar et al. 2022 — accurate post-training quantization for GPT models.

a9ab1ec12062f7ae · 2 sources · 100% confidence

LangChain framework publicly released on: 2022-10-25 by Harrison Chase.

c73295c7d563d226 · 2 sources · 100% confidence

LlamaIndex framework publicly released on: 2022-11-09 by Jerry Liu (originally GPT Index).

8ec0a4529b895d03 · 2 sources · 100% confidence

tiktoken publicly released on: 2022-12-06 by OpenAI — BPE tokenizer for OpenAI models.

70db729b5808e22c · 2 sources · 100% confidence

InstructGPT introduced in: Ouyang et al. 2022 — RLHF-tuned GPT-3, direct ancestor of ChatGPT.

590b9de765b8126e · 2 sources · 100% confidence

Speculative decoding introduced in: Leviathan, Kalman, Matias 2023 — Google Research.

6cdc7730bf41bb3d · 2 sources · 100% confidence

MTEB benchmark introduced in: Muennighoff et al. 2022 — Massive Text Embedding Benchmark.

cccd161dd058a31e · 2 sources · 100% confidence

Perplexity AI founded in: 2022 — AI-powered answer engine.

a4b4d4b7d5ab6740 · 2 sources · 100% confidence

ElevenLabs founded in: 2022 — AI voice synthesis platform.

dbdffd0d513ce369 · 2 sources · 95% confidence

Midjourney publicly released on: 2022-07-12 — public beta launch.

186ee5950b1f645e · 2 sources · 100% confidence

Anthropic Constitutional AI Harmlessness introduced in paper: Bai et al. 2022 — training a helpful and harmless assistant.

6fa575eb9df5ac32 · 2 sources · 100% confidence

ReAct prompting pattern introduced in: Yao et al. 2022 — synergizing reasoning and acting in language models.

95193a0b79c777e8 · 2 sources · 100% confidence

Chain-of-Thought (CoT) introduced in: Wei et al. 2022 — Chain-of-Thought Prompting Elicits Reasoning in Large Language Models.

a8503ad535423b54 · 2 sources · 100% confidence

Galactica publicly released on: 2022-11-15 by Meta AI — scientific LLM (withdrawn after 3 days).

badcfc0b4bb997f1 · 2 sources · 100% confidence

Fireworks AI founded in: 2022 — fast inference for open-source models.

c47d097cc3bceaaa · 2 sources · 95% confidence

HELM introduced in paper: Holistic Evaluation of Language Models (Liang et al., Stanford CRFM 2022-11-16).

494f2bf84f0e5dd2 · 2 sources · 100% confidence

Inflection AI founded in: 2022 by Mustafa Suleyman, Reid Hoffman, Karén Simonyan — consumer-AI company; creator of Pi assistant.

981662c18b08c628 · 2 sources · 100% confidence

Adept AI founded in: 2022 by David Luan + Niki Parmar + Ashish Vaswani — agent-AI company building action transformer models.

8e33113147c2f543 · 2 sources · 100% confidence

Langfuse founded in: 2022 by Marc Klingen + Max Deichmann + Clemens Rawert — open-source LLM observability + tracing platform.

e10249c02a470237 · 2 sources · 100% confidence

Helicone founded in: 2022 by Justin Torre + Cole Gottdank + Scott Nguyen — open-source LLM observability + analytics (YC W23).

a3fec2abf9eaadd8 · 2 sources · 100% confidence

Replit Ghostwriter publicly released on: 2022-10-26 by Replit — AI pair-programmer (renamed Replit AI in 2023).

6621efa92dc14702 · 2 sources · 100% confidence

Hugging Face Inference Endpoints publicly released on: 2022-09-27 by Hugging Face — managed inference deployment for any HF model.

87407b8052918e59 · 2 sources · 100% confidence

Brave Search API publicly released on: 2022-03-22 by Brave Software — independent search index API for AI grounding + privacy-focused search.

e40cdeb9b02f70a7 · 2 sources · 100% confidence

Recraft founded in: 2022 by Anna Veronika Dorogush — text-to-image platform with brand-style consistency + vector + 3D output.

f33bc9a6d1e921ad · 2 sources · 95% confidence

BIG-bench introduced in paper: Beyond the Imitation Game: Quantifying and extrapolating the capabilities of language models (Srivastava et al., 2022).

bde28f6f7e14e0e9 · 3 sources · 92% confidence