Skip to main content

Topic collection

Language Model APIs

Model reviews and comparisons covering context, tool use, coding workflows, pricing, and API access.

The language-model guides compare APIs by the work they need to perform, including structured output, tool use, coding, long-context processing, latency, and price. They avoid treating a single benchmark as a universal ranking. Use the articles to understand tradeoffs and the linked model pages to inspect current capabilities before selecting a model for an application or agent workflow.

14 articles

Articles about language models

GPT 6.2 Astra vs Fable 5.2 API: Speed and Price Tested (2026)
21 min read

GPT 6.2 Astra vs Fable 5.2 API: Speed and Price Tested (2026)

Astra vs Fable: GPT 6.2 Astra and Claude Fable 5.2 tied on 5 graded API tasks. Fable wrote 3.5x the tokens, finished sooner and billed 6.5x per call.

Read article
Claude Fable 5.2 vs Opus 5.5 API: Tested on Real Cost (2026)
12 min read

Claude Fable 5.2 vs Opus 5.5 API: Tested on Real Cost (2026)

Claude Fable 5.2 beats Opus 5.5 on 10 of 11 Anthropic benchmarks, but in our API tests both solved every task and Opus did it for a fraction of the cost.

Read article
Composer 2.5 API Pricing: Fast vs Standard Cost per 1M (2026)
12 min read

Composer 2.5 API Pricing: Fast vs Standard Cost per 1M (2026)

Composer 2.5 API pricing per 1M input and output tokens, Fast vs standard, what 100 to 10,000 agent runs cost a month, the benchmarks behind the rate, and code.

Read article
Gemini API Pricing: Argon, Flash, TTS per 1M Tokens (2026)
12 min read

Gemini API Pricing: Argon, Flash, TTS per 1M Tokens (2026)

Gemini API pricing per 1M tokens for 3.8 Flash and Gemini 4 Argon, Carbon and Hydrogen, Gemini TTS per second, Google's upcoming price changes, and real costs.

Read article
OpenAI API Pricing: GPT 6 Astra, Sol, Luna per 1M (2026)
12 min read

OpenAI API Pricing: GPT 6 Astra, Sol, Luna per 1M (2026)

OpenAI API pricing per 1M tokens for GPT 6 Astra, Sol, Luna and GPT 5: input, output, cache and long-context rates, real request costs, and which to pick.

Read article
Claude API Pricing: Anthropic Rates per 1M Tokens (2026)
12 min read

Claude API Pricing: Anthropic Rates per 1M Tokens (2026)

Anthropic API pricing per 1M input and output tokens for Claude Opus 5.5, Sonnet 5.5, Haiku 5.5 and Fable 5.2, with cache rates and real Messages API costs.

Read article
DeepSeek API Pricing: V4 Pro, V4 Flash per 1M Tokens (2026)
17 min read

DeepSeek API Pricing: V4 Pro, V4 Flash per 1M Tokens (2026)

DeepSeek pricing per 1M tokens for V4 Pro and V4.1 Flash, which replaced V4 Flash: weekday peak rates double off-peak, cache-hit costs, and how to call the API.

Read article
Gemini 3.7 Flash API Pricing & Cost per 1M Tokens (2026)
10 min read

Gemini 3.7 Flash API Pricing & Cost per 1M Tokens (2026)

Gemini 3.7 Flash API pricing per 1M tokens: Google deprecated it on October 8, 2026 and routes calls to 3.8 Flash. Our rate, benchmarks, and how to switch.

Read article
GPT-5.6 Sol Ultrafast API: 750 Tokens/sec on Cerebras (2026)
10 min read

GPT-5.6 Sol Ultrafast API: 750 Tokens/sec on Cerebras (2026)

OpenAI's Ultrafast mode ran GPT-5.6 Sol at up to 750 tokens/sec on Cerebras in an August preview. The public tier now covers only GPT-6.1 Sol and GPT-6 Astra.

Read article
Grok 4.6 API Pricing, SWE-bench 95.6%, Cost per 1M (2026)
13 min read

Grok 4.6 API Pricing, SWE-bench 95.6%, Cost per 1M (2026)

Grok 4.6 scores 95.6% on SWE-bench Verified (Vals AI, 4th) and 40.4% on CursorBench 4.0 at High. API pricing per 1M tokens, 500k context, how to call it.

Read article
Qwen3.8 27B: Benchmarks, Specs, and How to Run It (2026)
15 min read

Qwen3.8 27B: Benchmarks, Specs, and How to Run It (2026)

Qwen3.8 27B is a 27B open-weight VL model under Apache 2.0: 61.7% SWE-bench Pro, 262K context, image and video input. Benchmarks, VRAM math, how to run it.

Read article
Fable 5 vs Kimi K3: API Price, Benchmarks, Which Wins (2026)
10 min read

Fable 5 vs Kimi K3: API Price, Benchmarks, Which Wins (2026)

Kimi K3 costs less per token than Claude Fable 5, but Fable 5 leads Arena 1513 to 1493 and HLE 53.3 to 43.5. API pricing, benchmarks, successors, and code.

Read article
GPT 5.6 Sol vs Kimi K3: Price, Benchmarks, Which Wins (2026)
10 min read

GPT 5.6 Sol vs Kimi K3: Price, Benchmarks, Which Wins (2026)

GPT 5.6 Sol costs less per token than Kimi K3 and beats it on DeepSWE, 73% to 69% in 61 steps vs 98. API pricing, benchmarks, speed, and code to call both.

Read article
Kimi K3 API Pricing & Cost per 1M Tokens, Benchmarks (2026)
12 min read

Kimi K3 API Pricing & Cost per 1M Tokens, Benchmarks (2026)

Kimi K3 API pricing and cost per 1M input and output tokens, what a long agent run costs, how K3 benchmarks against Claude and GPT, and how to call it.

Read article