Code Models | Inference Hub

GPT OSS 120B

OpenAI

LLM

Open-weight 117B MoE model (5.1B active) achieving near o4-mini reasoning. Apache 2.0 licensed, runs on a single 80GB GPU.

117B MoE (5.1B active) 131k ctx Open

7 providers

$0.070 /MTok input

from Venice AI

DeepSeek V3.1

DeepSeek

LLM

Updated DeepSeek V3 with improved coding and reasoning performance.

671B MoE 128k ctx Open

7 providers

$0.150 /MTok input

from OpenRouter

Kimi K2

Moonshot AI

LLM

State-of-the-art 1T MoE model with 32B active parameters. Strong coding and agentic capabilities.

1T MoE (32B active) 128k ctx Open

9 providers

$0.195 /MTok input

from AIMLAPI

DeepSeek V3

DeepSeek

LLM

Open-weight 671B MoE model with strong coding and reasoning at low cost.

671B MoE 128k ctx Open

10 providers

$0.200 /MTok input

from OpenRouter

Kimi K2.5

Moonshot AI

LLM

Open-weight multimodal model with agent swarm mode supporting up to 100 parallel sub-agents.

128k ctx Open

12 providers

$0.230 /MTok input

from SiliconFlow

DeepSeek V3.2

DeepSeek

LLM

Latest DeepSeek V3 with improved reasoning and coding. 671B MoE (37B active), MIT licensed, 164K context.

671B MoE (37B active) 164k ctx Open

10 providers

$0.260 /MTok input

from OpenRouter

DeepSeek R1

DeepSeek

LLM

Reasoning-focused model with chain-of-thought capabilities rivaling o1.

671B MoE 128k ctx Open

8 providers

$0.280 /MTok input

from DeepSeek

DeepSeek R1 0528

DeepSeek

LLM

Updated R1 with improved reasoning accuracy and reduced hallucination.

671B MoE 128k ctx Open

8 providers

$0.450 /MTok input

from OpenRouter

GPT-5 Codex

OpenAI

LLM

GPT-5 variant optimized for code generation and software engineering.

128k ctx

2 providers

$0.500 /MTok input

from KIE AI

GPT-5.1 Codex

OpenAI

LLM

GPT-5.1 code-optimized variant.

128k ctx

2 providers

$0.500 /MTok input

from KIE AI

GPT-5.2 Codex

OpenAI

LLM

GPT-5.2 code-optimized variant.

128k ctx

2 providers

$0.700 /MTok input

from KIE AI

GPT-5.3 Codex

OpenAI

LLM

GPT-5.3 code-optimized variant.

128k ctx

2 providers

$0.700 /MTok input

from KIE AI

GPT-5.4

OpenAI

LLM

OpenAI's latest frontier model combining reasoning, coding, and agentic workflows.

128k ctx

7 providers

$0.700 /MTok input

from KIE AI

GPT-5.4 Codex

OpenAI

LLM

Latest GPT-5.4 code-optimized variant with industry-leading coding capabilities.

128k ctx

2 providers

$0.700 /MTok input

from KIE AI

GLM 5

Zhipu AI

LLM

Frontier 744B model trained on Huawei Ascend chips. Open source with strong agentic capabilities.

744B 128k ctx Open

11 providers

$0.720 /MTok input

from OpenRouter

Claude 4.6 Sonnet

Anthropic

LLM

Latest Sonnet with Opus-tier capabilities at Sonnet pricing.

200k ctx

9 providers

$1.05 /MTok input

from KIE AI

GLM 5.1

Z.ai

LLM

Coding-focused frontier model scoring 94% of Claude Opus 4.6. 744B MoE trained on Huawei Ascend 910B. #1 on SWE-Bench Pro (open source).

744B MoE (40B active) 203k ctx

3 providers

$1.40 /MTok input

from Fireworks AI

Claude 4.6 Opus

Anthropic

LLM

Anthropic's most capable model with 1M token context and advanced reasoning.

1M ctx

9 providers

$1.75 /MTok input

from KIE AI

o3

OpenAI

LLM

Reasoning-focused model with step-by-step deliberation for complex math, coding, and science tasks.

200k ctx

3 providers

$2.00 /MTok input

from OpenRouter

o3 Pro

OpenAI

LLM

Most capable reasoning model in OpenAI's lineup with extended thinking for maximum reliability.

200k ctx

2 providers

$20.00 /MTok input

from OpenRouter

GPT-5.4 Pro

OpenAI

LLM

Highest capability GPT-5.4 tier with maximum reasoning depth. Premium pricing.

128k ctx

3 providers

$30.00 /MTok input

from OpenRouter