Code
Browse models for code and compare pricing across providers.
GPT OSS 120B
OpenAIOpen-weight 117B MoE model (5.1B active) achieving near o4-mini reasoning. Apache 2.0 licensed, runs on a single 80GB GPU.
Qwen3 Coder 30B A3B Instruct
AlibabaQwen3 Coder 30B A3B Instruct, an open-weight Alibaba model available via OpenRouter.
Qwen3 Coder Next
AlibabaQwen3 Coder Next, an open-weight Alibaba model available via OpenRouter.
DeepSeek R1
DeepSeekReasoning-focused model with chain-of-thought capabilities rivaling o1.
DeepSeek V3.2
DeepSeekLatest DeepSeek V3 with improved reasoning and coding. 671B MoE (37B active), MIT licensed, 164K context.
Kimi K2
Moonshot AIState-of-the-art 1T MoE model with 32B active parameters. Strong coding and agentic capabilities.
Qwen3 Coder Flash
AlibabaQwen3 Coder Flash, an open-weight Alibaba model available via OpenRouter.
DeepSeek V3
DeepSeekOpen-weight 671B MoE model with strong coding and reasoning at low cost.
DeepSeek V3.1
DeepSeekUpdated DeepSeek V3 with improved coding and reasoning performance.
Qwen3 Coder 480B A35B
AlibabaQwen3 Coder 480B A35B, an open-weight Alibaba model available via OpenRouter.
GPT-5.1-Codex-Mini
OpenAIGPT-5.1-Codex-Mini, a proprietary OpenAI model available via OpenRouter.
Kimi K2.5
Moonshot AIOpen-weight multimodal model with agent swarm mode supporting up to 100 parallel sub-agents.
DeepSeek R1 0528
DeepSeekUpdated R1 with improved reasoning accuracy and reduced hallucination.
GPT-5 Codex
OpenAIGPT-5 variant optimized for code generation and software engineering.
GPT-5.1 Codex
OpenAIGPT-5.1 code-optimized variant.
GLM 5
Zhipu AIFrontier 744B model trained on Huawei Ascend chips. Open source with strong agentic capabilities.
Qwen3 Coder Plus
AlibabaQwen3 Coder Plus, an open-weight Alibaba model available via OpenRouter.
Kimi K2.6
Moonshot AIMoonshot AI's open-weight flagship: natively multimodal 1T MoE model (32B active) with text, image, and video input, thinking and non-thinking modes, and 256K context. State-of-the-art open-source coding and agentic performance (SWE-Bench Verified 80.2); Agent Swarm scales to 300 parallel sub-agents.
GPT-5.2 Codex
OpenAIGPT-5.2 code-optimized variant.
GPT-5.3 Codex
OpenAIGPT-5.3 code-optimized variant.
GPT-5.4
OpenAIOpenAI's latest frontier model combining reasoning, coding, and agentic workflows.
GPT-5.4 Codex
OpenAILatest GPT-5.4 code-optimized variant with industry-leading coding capabilities.
Kimi K2.7 Code
Moonshot AICoding-focused agentic model built on Kimi K2.6 with ~30% lower thinking-token usage. 1T MoE (32B active), native INT4 quantization, 256K context, text/image/video input. Forces thinking and preserve_thinking modes; tuned for long-horizon software engineering and agent workflows.
GLM 5.1
Z.aiCoding-focused frontier model scoring 94% of Claude Opus 4.6. 744B MoE trained on Huawei Ascend 910B. #1 on SWE-Bench Pro (open source).
Claude 4.6 Sonnet
AnthropicLatest Sonnet with Opus-tier capabilities at Sonnet pricing.
GPT-5.1-Codex-Max
OpenAIGPT-5.1-Codex-Max, a proprietary OpenAI model available via OpenRouter.
Claude 4.6 Opus
AnthropicAnthropic's most capable model with 1M token context and advanced reasoning.
o3
OpenAIReasoning-focused model with step-by-step deliberation for complex math, coding, and science tasks.
Fugu Ultra
Sakana AIThe high-tier variant of Sakana AI's Fugu orchestration model. Reportedly outperforms Opus 4.8, GPT-5.5 and Gemini 3.1 Pro on many benchmarks (73.7 on SWE-bench Pro), standing shoulder-to-shoulder with Claude Fable 5. Built on the TRINITY and Conductor (ICLR 2026) research.
o3 Pro
OpenAIMost capable reasoning model in OpenAI's lineup with extended thinking for maximum reliability.
GPT-5.4 Pro
OpenAIHighest capability GPT-5.4 tier with maximum reasoning depth. Premium pricing.