The Inference Report

November 6, 2026

AI Labs — News

The infrastructure layer is consolidating around agentic workflows, and the announcements reveal where vendors are placing their bets. OpenAI is positioning models as operational tools embedded directly into customer workflows, Cult.fit and Merge both showcase GPT-6 variants and Codex handling code review, business logic, and administrative work. NVIDIA's messaging centers on the CPU-GPU handoff problem: agents don't just reason, they execute across databases, search, and code generation, requiring orchestration that favors vendors who control the full stack. AWS amplified this by making Step Functions callable as agent tools, converting existing enterprise workflow infrastructure into AI-native infrastructure without requiring rewrites. AMD and NVIDIA are both racing to solve memory constraints through quantization and codesign approaches, AMD's online quantization and NVIDIA's HBM4 adoption signal that model serving economics are tightening and that vendors who can shrink memory footprint while preserving performance will capture margin in the agent-heavy deployment phase ahead.

The secondary pattern is infrastructure vendors claiming territory in specialized domains while maintaining agnosticism on model choice. Google DeepMind's Gemini Omni 1.2 Flash emphasizes speed and cost efficiency; Hugging Face is distributing NVIDIA Cosmos 3 for world simulation and Phonon TTS for on-device speech; Mistral's Studio is positioning Le Chonk as inference-ready; Anthropic continues the scientific discovery narrative with Claude. The message is consistent: models are commoditizing faster than expected, and differentiation is shifting to orchestration, quantization, and domain-specific applications. IBM's quantum error correction initiative with Oak Ridge and AMD is the outlier, a longer-term bet on quantum infrastructure that doesn't compete with today's agentic deployments but signals where serious compute investment is hedging. OpenAI's National Community Compact and Compute Campus Policy Blueprint, meanwhile, reads as infrastructure positioning dressed in civic language, securing compute supply and favorable regulatory treatment before the infrastructure bottleneck becomes acute.

Sloane Duvall

AMD

AWS

AWS Weekly Roundup: GPT-6.1 Terra on Bedrock, Step Functions as agent tools, and more (November 2, 2026)

Anthropic

AI for scientific discovery: How researchers use Claude

Google

Detecting signs of hypertension using wearable sensors

Google DeepMind

Gemini Omni 1.2 Flash: Fast and cost-effective with longer clips

Hugging Face

IBM

Oak Ridge National Lab, IBM, and AMD Launch New Co-Design Platform for Quantum Error Correction

Liquid AI

Mistral

Le Chonk brings the brains. Studio puts it to work.

NVIDIA

OpenAI

AI Labs — Models

A curated reference of models from major AI labs, with open/closed weight status, input modalities, and context window size. American labs tend towards closed weights models and Chinese labs tend toward open weights models.

AllenAI

Closed Weights

None

Open Weights

AllenAI: Olmo 3 32B Think
Text66K

Amazon

Closed Weights

Amazon: Nova 2 Lite
TextVisionVideoFiles1M

Open Weights

None

Anthropic

Closed Weights

Anthropic: Claude Haiku 5.5
TextVisionFiles1M
Anthropic: Claude Opus 4.5
FilesVisionText200K
Anthropic: Claude Opus 4.6
TextVisionFiles1M
Anthropic: Claude Opus 4.7
TextVisionFiles1M
Anthropic: Claude Opus 4.8
TextVisionFiles1M
Anthropic: Claude Opus 5
TextVisionFiles1M
Anthropic: Claude Opus 5.5
TextVisionFiles1M
Anthropic: Claude Sonnet 4.6
TextVisionFiles1M
Anthropic: Claude Sonnet 5
TextVisionFiles1M
Anthropic: Claude Sonnet 5.5
TextVisionFiles1M

Open Weights

None

Arcee AI

Closed Weights

None

Open Weights

Arcee AI: Trinity Large Thinking
Text262K

Deep Cogito

Closed Weights

None

Open Weights

Deep Cogito: Cogito v2.1 671B
Text128K

Google

Closed Weights

Google: Gemini 3 Flash Preview
TextVisionFilesAudioVideo1M
Google: Gemini 3.1 Flash Lite
TextVisionVideoFilesAudio1M
Google: Gemini 3.1 Pro Preview
AudioFilesVisionTextVideo1M
Google: Gemini 3.5 Flash
TextVisionVideoFilesAudio1M
Google: Gemini 3.5 Flash Lite
TextVisionVideoFilesAudio1M
Google: Gemini 3.6 Flash
TextVisionVideoFilesAudio1M
Google: Gemini 3.7 Flash
TextVisionVideoFilesAudio1M
Google: Gemini 3.8 Flash
TextVisionVideoFilesAudio1M
Google: Gemini 4 Argon
AudioFilesVisionTextVideo1M
Google: Nano Banana 2 (Gemini 3.1 Flash Image)
VisionText131K
Google: Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image)
VisionText66K
Google: Nano Banana Pro (Gemini 3 Pro Image)
VisionText66K

Open Weights

Google: Gemma 4 26B A4B
VisionTextVideo262K
Google: Gemma 4 31B
VisionTextVideo262K

IBM

Closed Weights

None

Open Weights

IBM: Granite 4.2 8B
Text131K

Inception

Closed Weights

Inception: Mercury 2
Text128K
Inception: Mercury 2.5
Text260K
Inception: Mercury 3
Text1M

Open Weights

None

Meta (Muse)

Closed Weights

None

Open Weights

Meta: Muse Glimmer 30B
TextVision131K
Meta: Muse Spark Open
TextVisionVideoFilesAudio1M

NVIDIA

Closed Weights

None

Open Weights

NVIDIA: Nemotron 3 Nano 30B A3B
Text262K
NVIDIA: Nemotron 3 Super
Text1M
NVIDIA: Nemotron 3 Ultra
Text262K
NVIDIA: Nemotron 3.5 Lightning
Text262K
NVIDIA: Nemotron 3.5 Super VL
TextVisionVideo1M
NVIDIA: Switchyard
Text1M

OpenAI

Closed Weights

OpenAI: GPT-5.1
VisionTextFiles400K
OpenAI: GPT-5.1-Codex
TextVision400K
OpenAI: GPT-5.1-Codex-Max
TextVision400K
OpenAI: GPT-5.1-Codex-Mini
VisionText400K
OpenAI: GPT-5.2
FilesVisionText400K
OpenAI: GPT-5.2-Codex
TextVision400K
OpenAI: GPT-5.3-Codex
TextVisionFiles400K
OpenAI: GPT-5.4
TextVisionFiles1M
OpenAI: GPT-5.4 Mini
FilesVisionText400K
OpenAI: GPT-5.4 Nano
FilesVisionText400K
OpenAI: GPT-5.4 Pro
TextVisionFiles1M
OpenAI: GPT-5.5
FilesVisionText1M
OpenAI: GPT-5.6 Luna
FilesVisionText1M
OpenAI: GPT-5.6 Luna Pro
FilesVisionText1M
OpenAI: GPT-5.6 Sol
FilesVisionText1M
OpenAI: GPT-5.6 Sol Pro
FilesVisionText1M
OpenAI: GPT-5.6 Terra
FilesVisionText1M
OpenAI: GPT-5.6 Terra Pro
FilesVisionText1M
OpenAI: GPT-6 Astra
FilesVisionText1M
OpenAI: GPT-6 Astra Pro
FilesVisionText1M
OpenAI: GPT-6 Luna
FilesVisionText1M
OpenAI: GPT-6 Luna Pro
FilesVisionText1M
OpenAI: GPT-6 Sol
FilesVisionText1M
OpenAI: GPT-6 Sol Pro
FilesVisionText1M
OpenAI: GPT-6.1 Sol
FilesVisionText1M
OpenAI: GPT-6.1 Sol Pro
FilesVisionText1M
OpenAI: GPT-6.1 Terra
FilesVisionText1M
OpenAI: GPT-6.1 Terra Pro
FilesVisionText1M

Open Weights

None

Perceptron

Closed Weights

Perceptron: Perceptron Mk1
TextVisionVideo33K
Perceptron: Perceptron Mk1.5
TextVisionVideoAudio37K

Open Weights

None

Poolside

Closed Weights

None

Open Weights

Poolside: Laguna S 2.1
Text1M
Poolside: Laguna XS 2.1
Text262K

Reka AI

Closed Weights

None

Open Weights

Reka Edge
VisionTextVideo16K

Relace

Closed Weights

None

Open Weights

Relace: Relace Search
Text256K

Thinking Machines

Closed Weights

None

Open Weights

Thinking Machines: Inkling
TextVisionAudio524K
Thinking Machines: Inkling Small
TextVisionAudio524K

xAI

Closed Weights

SpaceXAI: Grok 4.20
TextVisionFiles2M
SpaceXAI: Grok 4.20 Multi-Agent
TextVisionFiles2M
SpaceXAI: Grok 4.3
TextVisionFiles1M
SpaceXAI: Grok 4.5
TextVisionFiles500K
SpaceXAI: Grok 4.6
TextVisionFiles500K
SpaceXAI: Grok 4.7
TextVisionFiles500K
SpaceXAI: Grok Build 0.1
TextVisionFiles256K

Open Weights

None

Cohere

Closed Weights

Cohere: Command A+
TextVision192K

Open Weights

None

Mistral AI

Closed Weights

None

Open Weights

Mistral: Devstral 2 2512
TextFiles262K
Mistral: Ministral 3 14B 2512
TextVision262K
Mistral: Ministral 3 3B 2512
TextVision131K
Mistral: Ministral 3 8B 2512
TextVision262K
Mistral: Mistral Large 3 2512
TextVisionFiles262K
Mistral: Mistral Large 4
TextVision1M
Mistral: Mistral Medium 3.5
TextVisionFiles262K
Mistral: Mistral Small 4
TextVision262K

Alibaba (Qwen)

Closed Weights

None

Open Weights

Qwen: Qwen3 Coder Next
Text262K
Qwen: Qwen3.5 397B A17B
TextVisionVideo262K
Qwen: Qwen3.5-122B-A10B
TextVisionVideo262K
Qwen: Qwen3.5-27B
TextVisionVideo262K
Qwen: Qwen3.5-35B-A3B
TextVisionVideo262K
Qwen: Qwen3.5-9B
TextVisionVideo262K
Qwen: Qwen3.6 27B
TextVisionVideo262K
Qwen: Qwen3.6 35B A3B
TextVisionVideo262K
Qwen: Qwen3.8 2.4T A95B
Text1M
Qwen: Qwen3.8 27B
TextVisionVideo1M

ByteDance

Closed Weights

Seed: Seed 1.6
VisionTextVideo262K
Seed: Seed 1.6 Flash
VisionTextVideo262K
Seed: Seed 2.1 Turbo
TextVisionVideo262K
Seed: Seed-2.0-Code
TextVisionVideo262K
Seed: Seed-2.0-Lite
TextVisionVideo262K
Seed: Seed-2.0-Mini
TextVisionVideo262K

Open Weights

None

DeepSeek

Closed Weights

None

Open Weights

DeepSeek: DeepSeek V3.2
Text164K
DeepSeek: DeepSeek V4 Flash 0423
Text1M
DeepSeek: DeepSeek V4 Flash 0731
Text1M
DeepSeek: DeepSeek V4 Flash Vision Exp
TextVision1M
DeepSeek: DeepSeek V4 Pro 0423
Text1M
DeepSeek: DeepSeek V4 Pro 0813
Text1M
DeepSeek: DeepSeek V4.1 Flash
TextVision1M
DeepSeek: DeepSeek V4.1 Flash 1017
TextVision1M

MiniMax

Closed Weights

None

Open Weights

MiniMax: MiniMax M2.5
Text198K
MiniMax: MiniMax M2.7
Text197K
MiniMax: MiniMax M3
TextVisionVideo1M
MiniMax: MiniMax M3.1 Flash
TextVisionVideo1M

Moonshot AI

Closed Weights

None

Open Weights

MoonshotAI: Kimi K2.5
TextVision262K
MoonshotAI: Kimi K2.6
TextVision262K
MoonshotAI: Kimi K2.7 Code
TextVision262K
MoonshotAI: Kimi K3
TextVisionVideo1M
MoonshotAI: Kimi K3.1 Pro
TextVisionVideo1M

Xiaomi

Closed Weights

None

Open Weights

Xiaomi: MiMo-V2.5
TextAudioVisionVideo1M
Xiaomi: MiMo-V2.5-Pro
Text262K
Xiaomi: MiMo-V2.6-Flash
TextVisionVideoAudio1M
Xiaomi: MiMo-V2.6-Pro
TextVisionVideoAudio1M

Z.ai

Closed Weights

None

Open Weights

Z.ai: GLM 4.6V
VisionTextVideo131K
Z.ai: GLM 4.7
Text205K
Z.ai: GLM 4.7 Flash
Text200K
Z.ai: GLM 5
Text205K
Z.ai: GLM 5 Turbo
Text203K
Z.ai: GLM 5.1
Text205K
Z.ai: GLM 5.2
Text1M
Z.ai: GLM 5.3
Text1M
Z.ai: GLM 5.3 Flash
TextVisionVideo1M
Z.ai: GLM 5.3 FlashX
TextVisionVideo1M
Z.ai: GLM 5.3 FlashX Legal
TextVisionVideo1M
Z.ai: GLM 5V Turbo
VisionTextVideo203K

Upstage

Closed Weights

Upstage: Solar Mini 4
Text524K
Upstage: Solar Pro 4
Text524K

Open Weights

None