ModelBench
Model catalog

Models

Every model in the catalog with source-linked pricing, context limits, provider availability, and published benchmark results.

Compare Export CSV
7 models
Structured output fireworks-ai
Ranked by Terminal-Bench 2.1
Moonshot AI: MoonshotAI: Kimi K3 T
Reasoning Tools JSON Open weights 88.3

Kimi K3 is a 2.8T parameter open-weight multimodal reasoning model from Moonshot AI. It is suited for complex coding, knowledge work, and long-horizon agentic workflows, and is particularly strong at...

moonshotai/kimi-k3 2026-07-16 1.04858M context $2.303/M input $11.55/M output
63 providers
DeepSeek: DeepSeek: DeepSeek V4 Flash 0731 T
Reasoning Tools JSON Open weights 82.7

DeepSeek V4 Flash 0731 is a sparse mixture-of-experts model from DeepSeek, with 13B active parameters out of 284B total. This re-post-trained revision is suited for coding, reasoning, and agent workflows....

deepseek/deepseek-v4-flash-0731 2026-07-31 1.04858M context $0.04/M input $0.08/M output
43 providers
MiniMax: MiniMax-M3 T
Reasoning Tools Open weights 66.0

MiniMax multimodal model for long-context coding, perception, and agent planning

minimax/MiniMax-M3 2026-06-01 1.04858M context $0.3/M input $1.2/M output
37 providers
DeepSeek: DeepSeek: DeepSeek V4 Pro 0423 T
Reasoning Tools JSON Open weights 64.7

DeepSeek V4 Pro is a large-scale Mixture-of-Experts model from DeepSeek with 1.6T total parameters and 49B activated parameters, supporting a 1M-token context window. It is designed for advanced reasoning, coding,...

deepseek/deepseek-v4-pro 2026-04-24 1.024M context $0.809/M input $1.619/M output
62 providers
NVIDIA: NVIDIA: Nemotron 3 Ultra T
Reasoning Tools Open weights 56.4

NVIDIA Nemotron 3 Ultra is an open frontier-reasoning and orchestration model from NVIDIA, with 55B active parameters out of 550B total (MoE). Built on a hybrid Transformer-Mamba mixture-of-experts architecture, it...

nvidia/nemotron-3-ultra-550b-a55b 2026-06-04 256K context $0.625/M input $3.125/M output
15 providers
Meta: Meta: Muse Glimmer 30B T
Reasoning Tools JSON Open weights 51.7

Muse Glimmer 30B is a dense, open-weight multimodal model from Meta Superintelligence Labs, distilled from Muse Spark and optimized for autonomous agents on consumer hardware. It is suited for long-horizon...

meta/muse-glimmer-30b 2026-08-10 131.072K context $0.3/M input $1.1/M output
9 providers
MiniMax: MiniMax-M2.7 T
Reasoning Tools Open weights 51.1

Open MiniMax flagship for coding agents, office automation, and complex environments

minimax/MiniMax-M2.7 2026-03-18 204.8K context $0.3/M input $1.2/M output
35 providers