Anthropic
Claude Opus 4.8 (Fast)
Fast-mode variant of Opus 4.8 - identical capabilities with higher output speed at 2x pricing relative to regular Opus 4.8.
ENTERPRISE DNA / MODEL GUIDE
Make sense of the AI model landscape. Find your shortlist, compare the tradeoffs and leave with a workflow you can use.
START WITH THE JOB
A shortlist. A starter prompt. A way to check the result.
2 new models, 8 price moves since yesterday.
DaisyRose 27B just landed at $0.15/M in; Gemma 3 4B dropped 50% on price.
UNDERSTAND THE LANDSCAPE
Start with what you want to produce. Open a group, explore a model and try its checkable exercise.
Complex repository changes
text + image + file → textClaude Sonnet 5.5Scoped features and debugging
text + image + file → textClaude Haiku 5.5High-volume classification
text + image + file → textClaude Fable 5.1Long-running planning and code review
text + image + file → textGPT-6.1 SolProfessional document and code work
file + image + text → textGPT-6.1 Sol ProHigher-effort variants of the same task
file + image + text → textGPT-6 LunaCost-conscious everyday automation
file + image + text → textGPT-6 AstraDemanding analysis and engineering
file + image + text → textGemini 4 ArgonMultimodal reasoning with long inputs
audio + file + image + text + video → textGemini 3.8 FlashMultimodal task assistance
text + image + video + file + audio → textGemini 3.5 Flash LiteFocused subagent work
text + image + video + file + audio → textDeepSeek V4.1 Flash 1017Dated DeepSeek development snapshot
text + image → textDeepSeek V4 Pro 0423Text-first reasoning and coding
text → textQwen3.8 Max PrimeHigher-throughput Qwen variant
text + image + video → textQwen3.8 Omni FlashAudio-video understanding
text + image + audio + video → textKimi K3Multimodal knowledge and code work
text + image + video → textMistral Large 4Multimodal engineering workflows
text + image → textGrok 4.7Reasoning and tool-connected analysis
text + image + file → text30 profiles with sourced specifications, checked 2026-10-20. Exercises are prepared, not evaluated model results. Inspect our evaluation method →
THE MODEL GUIDE
24 models loaded
Shortlists use listed capabilities and published benchmarks, not Enterprise DNA task tests. Missing scores appear as “Not measured”. API prices are separate from app subscriptions.
Anthropic
Fast-mode variant of Opus 4.8 - identical capabilities with higher output speed at 2x pricing relative to regular Opus 4.8.
Anthropic
Fast-mode variant of Opus 4.7 - identical capabilities with higher output speed at premium 6x pricing.
Flagship Claude model for deep reasoning, coding, and long-horizon agents.
Anthropic
Everyday Claude agent model for coding, planning, browsing, and general work.
Accounts
Flagship GLM model for hybrid reasoning, coding, and agentic engineering.
Cf
Open flagship GLM for long-horizon coding agents and million-token context work.
Hf Zai Org
Open flagship GLM for long-horizon coding agents and million-token context work.
Subconscious
Open flagship GLM for long-horizon coding agents and million-token context work.
Zai Org
Open flagship GLM for long-horizon coding agents and million-token context work.
Balanced Claude model for coding, analysis, agent workflows, and cost control.
Accounts
MiniMax multimodal coding model for long-context reasoning and agent tasks.
Hf Minimaxai
MiniMax multimodal model for long-context coding, perception, and agent planning.
Accounts
Open MoE flagship with million-token context for coding and long agent runs.
Flagship DeepSeek model for coding, reasoning, and agentic work.
DeepSeek
Flagship DeepSeek model for coding, reasoning, and agentic work.
OpenAI
GPT-6 Sol is the cost-efficient high-end model in OpenAI's GPT-6 series, positioned below the flagship GPT-6 Astra and above the fast GPT-6 Luna tier.
OpenAI
GPT-6 Sol is the cost-efficient high-end model in OpenAI's GPT-6 series, positioned below the flagship GPT-6 Astra and above the fast GPT-6 Luna tier.
Accounts
Kimi reasoning model for long-horizon research, planning, and tool use.
Workers Ai
Kimi multimodal agent model for visual understanding, coding, and planning.
Novita
Kimi multimodal agent model for visual understanding, coding, and planning.
Kimi multimodal agent model for visual understanding, coding, and planning.
Flagship Claude model for deep reasoning, coding, and long-horizon agents.
Accounts
Kimi coding model for software agents, refactors, and repository reasoning.
Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking.
No models match those filters. Try another provider or reset your filters.
YOUR SHORTLIST
Estimate API cost per task
USD estimate = input tokens × input rate + output tokens × output rate, divided by 1 million. Excludes caching, tools, retries, extra reasoning tokens and provider fees. Check context and output limits before running.
FROM CHOOSING TO DOING